NVIDIA Windows GPU Display Driver contains a vulnerability in the DirectX 10 Usermode driver, where a specially crafted pixel shader can cause writing to unallocated memory, leading to denial of service or potential code execution.

Some NVIDIA Tegra mobile processors released prior to 2016 contain a buffer overflow vulnerability in BootROM Recovery Mode (RCM). An attacker with physical access to the device's USB and the ability to force the device to reboot into RCM could exploit the vulnerability to execute unverified code.

NVIDIA GPU Display Driver for Windows and Linux contains a vulnerability in the kernel mode layer handler, where an out-of-bounds access may lead to denial of service or data tampering.

NVIDIA DGX A100 SBIOS contains a vulnerability where an attacker may modify arbitrary memory of SMRAM by exploiting the GenericSio and LegacySmmSredir SMM APIs. A successful exploit of this vulnerability may lead to denial of service, escalation of privileges, and information disclosure.

NVIDIA DGX A100 SBIOS contains a vulnerability where an attacker may modify arbitrary memory of SMRAM by exploiting the NVME SMM API. A successful exploit of this vulnerability may lead to denial of service, escalation of privileges, and information disclosure.

A heap buffer overflow was discovered in the device control ioctl in the Linux driver for Nvidia graphics cards, which may allow an attacker to overflow 49 bytes. This issue was fixed in version 295.53.

NVIDIA DGX Spark GB10 contains a vulnerability in SROOT firmware, where an attacker could cause unexpected memory buffer operations. A successful exploit of this vulnerability might lead to data tampering, denial of service, or escalation of privileges.

All versions of NVIDIA Windows GPU Display Driver contain a vulnerability in the kernel mode layer (nvlddmkm.sys) handler for DxgkDdiEscape where untrusted input is used for buffer size calculation leading to denial of service or escalation of privileges.

NVIDIA Windows GPU Display Driver contains a vulnerability in the kernel mode layer helper function where an incorrect calculation of string length may lead to denial of service.

NVIDIA Shield TV Experience prior to v8.0.1, NVIDIA Tegra bootloader contains a vulnerability where the software performs an incorrect bounds check, which may lead to buffer overflow resulting in escalation of privileges and code execution. escalation of privileges, and information disclosure, code execution, denial of service, or escalation of privileges.

NVIDIA BMC contains a vulnerability in SPX REST API, where an authorized attacker can read and write to arbitrary locations within the memory context of the IPMI server process, which may lead to code execution, denial of service, information disclosure and data tampering.

DGX A100 SBIOS contains a vulnerability in Bds, which may lead to code execution, denial of service, or escalation of privileges.

NVIDIA GPU Display Driver for Linux contains a vulnerability in the kernel mode layer, where an unprivileged regular user can cause the use of an out-of-range pointer offset, which may lead to data tampering, data loss, information disclosure, or denial of service.

An elevation of privilege vulnerability in the Direct rendering infrastructure of the NVIDIA Tegra X1 where an unchecked input from userspace is passed as a pointer to kfree. This could lead to kernel memory corruption and possible code execution. This issue is rated as moderate. Product: Pixel. Version: N/A. Android ID: A-38415808. References: N-CVE-2017-0866.

All versions of NVIDIA Windows GPU Display Driver contain a vulnerability in the kernel mode layer (nvlddmkm.sys) implementation of the SubmitCommandVirtual DDI (DxgkDdiSubmitCommandVirtual) where untrusted input is used to reference memory outside of the intended boundary of the buffer leading to denial of service or escalation of privileges.

All versions of NVIDIA Windows GPU Display Driver contain a vulnerability in the kernel mode layer (nvlddmkm.sys) handler for DxgkDdiEscape where the size of an input buffer is not validated, leading to denial of service or potential escalation of privileges.

All versions of NVIDIA Windows GPU Display Driver contain a vulnerability in the kernel mode layer (nvlddmkm.sys) handler for DxgDdiEscape where the size of an input buffer is not validated, leading to denial of service or potential escalation of privileges.

For the NVIDIA Quadro, NVS, and GeForce products, NVIDIA GeForce Experience R340 before GFE 2.11.4.125 and R375 before GFE 3.1.0.52 contains a vulnerability in the kernel mode layer (nvstreamkms.sys) allowing a user to cause a stack buffer overflow with specially crafted executable paths, leading to a denial of service or escalation of privileges.

All versions of NVIDIA Windows GPU Display Driver contain a vulnerability in the kernel mode layer handler for DxgDdiEscape where the size of an input buffer is not validated leading to a denial of service or possible escalation of privileges

Stack-based buffer overflow in nvhost_job.c in the NVIDIA video driver for Android, Shield TV before OTA 3.3, Shield Table before OTA 4.4, and Shield Table TK1 before OTA 1.5.

NVIDIA Jetson Linux Driver Package contains a vulnerability in the Cboot module tegrabl_cbo.c, where, if TFTP is enabled, a local attacker with elevated privileges can cause a memory buffer overflow, which may lead to code execution, loss of Integrity, limited denial of service, and some impact to confidentiality.

Buffer overflow in nvhost_job.c in the NVIDIA video driver for Android, Shield TV before OTA 3.3, Shield Table before OTA 4.4, and Shield Table TK1 before OTA 1.5.

Buffer overflow in the NVIDIA GPU driver before 304.88, 310.x before 310.44, and 313.x before 313.30 for the X Window System on UNIX, when NoScanout mode is enabled, allows remote authenticated users to execute arbitrary code via a large ARGB cursor.

NVIDIA Windows GPU Display Driver contains a vulnerability in the kernel mode layer handler for DxgkDdiEscape in which the software uses a sequential operation to read from or write to a buffer, but it uses an incorrect length value that causes it to access memory that is outside of the bounds of the buffer which may lead to denial of service, escalation of privileges, code execution or information disclosure.

NVIDIA Windows GPU Display Driver contains a vulnerability in the kernel mode layer handler for DxgkDdiEscape in which the software uses a sequential operation to read from or write to a buffer, but it uses an incorrect length value that causes it to access memory that is outside of the bounds of the buffer, which may lead to denial of service or escalation of privileges.

The Escape interface in the Kernel Mode Driver layer in the NVIDIA GPU graphics driver R340 before 341.95 and R352 before 354.74 on Windows allows local users to obtain sensitive information, cause a denial of service (crash), or gain privileges via unspecified vectors related to an untrusted pointer, which trigger uninitialized or out-of-bounds memory access.

Unspecified vulnerability in the NVAPI support layer in the NVIDIA GPU graphics driver R340 before 341.92, R352 before 354.35, and R358 before 358.87 on Windows allows local users to obtain sensitive information, cause a denial of service (crash), or possibly gain privileges via unknown vectors. NOTE: this identifier was SPLIT from CVE-2015-7869 per ADT2 and ADT3 due to different vulnerability types and affected versions.

NVIDIA GPU Display Driver for Linux contains a vulnerability in the kernel mode layer, where improper restriction of operations within the bounds of a memory buffer can lead to denial of service, information disclosure, and data tampering.

NVIDIA GPU Display Driver for Windows and Linux contains a vulnerability in the kernel mode layer handler, where an unprivileged user can cause improper restriction of operations within the bounds of a memory buffer cause an out-of-bounds read, which may lead to denial of service.

NVIDIA Triton Inference Server for Linux contains a vulnerability in shared memory APIs, where a user can cause an improper memory access issue by a network API. A successful exploit of this vulnerability might lead to denial of service and data tampering.

The NVIDIA driver before 307.78, and Release 310 before 311.00, in the NVIDIA Display Driver service on Windows does not properly handle exceptions, which allows local users to gain privileges or cause a denial of service (memory overwrite) via a crafted application.

STDU Viewer 1.6.375 allows attackers to execute arbitrary code or cause a denial of service via a crafted .xps file, related to a "Read Access Violation on Control Flow starting at Unknown Symbol @ 0x0000000003aa7cef called from Unknown Symbol @ 0x0000000004aa024d."

TensorFlow is an end-to-end open source platform for machine learning. The implementation of `tf.raw_ops.MaxPool3DGradGrad` is vulnerable to a heap buffer overflow. The implementation(https://github.com/tensorflow/tensorflow/blob/596c05a159b6fbb9e39ca10b3f7753b7244fa1e9/tensorflow/core/kernels/pooling_ops_3d.cc#L694-L696) does not check that the initialization of `Pool3dParameters` completes successfully. Since the constructor(https://github.com/tensorflow/tensorflow/blob/596c05a159b6fbb9e39ca10b3f7753b7244fa1e9/tensorflow/core/kernels/pooling_ops_3d.cc#L48-L88) uses `OP_REQUIRES` to validate conditions, the first assertion that fails interrupts the initialization of `params`, making it contain invalid data. In turn, this might cause a heap buffer overflow, depending on default initialized values. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

mar_read.c in the Updater in Mozilla Firefox before 40.0 and Firefox ESR 38.x before 38.2 allows local users to gain privileges or cause a denial of service (out-of-bounds write) via a crafted name of a Mozilla Archive (aka MAR) file.

In display drm, there is a possible memory corruption due to a missing bounds check. This could lead to local escalation of privilege with System execution privileges needed. User interaction is not needed for exploitation. Patch ID: ALPS07780685; Issue ID: ALPS07780685.

A kernel pool overflow in the driver hitmanpro37.sys in Sophos SurfRight HitmanPro before 3.7.20 Build 286 (included in the HitmanPro.Alert solution and Sophos Clean) allows local users to escalate privileges via a malformed IOCTL call.

A flaw was found in libcaca v0.99.beta19. A buffer overflow issue in caca_resize function in libcaca/caca/canvas.c may lead to local execution of arbitrary code in the user context.

Stack-based buffer overflow in the condor_ schedd daemon in Condor before 7.0.5 allows attackers to cause a denial of service (crash) and possibly execute arbitrary code via unknown vectors.

Stack-based buffer overflow in the libbecompat library in Ingres 2.6, Ingres 2006 release 1 (aka 9.0.4), and Ingres 2006 release 2 (aka 9.1.0) on Linux and HP-UX allows local users to gain privileges by setting a long value of an environment variable before running (1) verifydb, (2) iimerge, or (3) csreport.

In Android for MSM, Firefox OS for MSM, QRD Android, with all Android releases from CAF using the Linux kernel, in the fastrpc kernel driver, a buffer overflow vulnerability from userspace may potentially exist.

In Qualcomm Android for MSM, Firefox OS for MSM, and QRD Android with all Android releases from CAF using the Linux kernel before security patch level 2018-04-05, the camera application triggers "user-memory-access" issue as the Camera CPP module Linux driver directly accesses the application provided buffer, which resides in user space. An unchecked userspace value (ioctl_ptr->len) is used to copy contents to a kernel buffer which can lead to kernel buffer overflow.

Buffer overflow in src/openttd.cpp in OpenTTD before 0.6.2 allows local users to execute arbitrary code via a large filename supplied to the "-g" parameter in the ttd_main function. NOTE: it is unlikely that this issue would cross privilege boundaries in typical environments.

In all android releases(Android for MSM, Firefox OS for MSM, QRD Android) from CAF using the linux kernel, Userspace can pass IEs to the host driver and if multiple append commands are received, then the integer variable that stores the length can overflow and the subsequent copy of the IE data may potentially lead to a heap buffer overflow.

Buffer overflow in errpt in IBM AIX 5.2, 5.3, and 6.1 allows local users to gain privileges via unknown attack vectors.

In Android for MSM, Firefox OS for MSM, QRD Android, with all Android releases from CAF using the Linux kernel, by calling an IPA ioctl and searching for routing/filer/hdr rule handle from ipa_idr pointer using ipa_idr_find() function, the wrong structure pointer can be returned resulting in a slab out of bound access in the IPA driver.

In Android for MSM, Firefox OS for MSM, QRD Android, with all Android releases from CAF using the Linux kernel, while processing the QCA_NL80211_VENDOR_SUBCMD_SET_TXPOWER_SCALE vendor command, in which attribute QCA_WLAN_VENDOR_ATTR_TXPOWER_SCALE contains fewer than 1 byte, a buffer overrun occurs.

STDU Viewer 1.6.375 allows attackers to execute arbitrary code or cause a denial of service via a crafted .jb2 file, related to a "User Mode Write AV starting at STDUJBIG2File!DllUnregisterServer+0x000000000000566e."

STDU Viewer 1.6.375 allows attackers to cause a denial of service or possibly have unspecified other impact via a crafted .jb2 file, related to "Data from Faulting Address controls Branch Selection starting at STDUJBIG2File!DllUnregisterServer+0x0000000000005578."

STDU Viewer 1.6.375 allows attackers to execute arbitrary code or cause a denial of service via a crafted .jb2 file, related to "Data from Faulting Address controls subsequent Write Address starting at STDUJBIG2File!DllGetClassObject+0x00000000000043e6."

TensorFlow is an end-to-end open source platform for machine learning. The implementation of `tf.raw_ops.AvgPool3DGrad` is vulnerable to a heap buffer overflow. The implementation(https://github.com/tensorflow/tensorflow/blob/d80ffba9702dc19d1fac74fc4b766b3fa1ee976b/tensorflow/core/kernels/pooling_ops_3d.cc#L376-L450) assumes that the `orig_input_shape` and `grad` tensors have similar first and last dimensions but does not check that this assumption is validated. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.