In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
IntheLinux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
Inthe Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In theLinux kernel, the following vulnerability has been resolved:drm/vmwgfx:Fix a deadlock in dma buffence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pendinglist, and thus doesn't require a locktofix poll->fence wait->fenceunref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and updatetheir status, to do that itholds a lock to preventthe list modifcations fromother threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, forwhich itholds a lock.dma buf polling cb unrefs afenceafter it's been signaled: so the pollcalls the wait,which signals the fences,which arebeing destroyed.The destruction tries to acquire the lock on the pending fences listwhich it cannever get because it's held by the wait fromwhichitwas called.Old bug, butnot a lot of userspace apps wereusing dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.TheLinux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
Inthe Linuxkernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fencepollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list,and thus doesn't require a lock tofixpoll->fence wait->fence unrefdeadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update theirstatus, to do that it holdsa lock to preventthe list modifcations from otherthreads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for whichit holdsa lock.dma buf polling cb unrefs a fenceafterit's been signaled: so the pollcalls the wait, whichsignals the fences, whichare beingdestroyed.The destruction tries to acquire the lock on the pending fences listwhich it can neverget because it's held by the wait from whichitwascalled.Old bug, but not alot of userspace apps were usingdma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linuxkernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
Inthe Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix adeadlock in dma buf fence pollingIntroduce aversion of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require alock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds alock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds alock.dma buf polling cb unrefs afence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not alot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fixa deadlock in dma buf fence pollingIntroducea version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't requirea lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holdsa lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holdsa lock.dma buf polling cb unrefsa fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but nota lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
| linux | | https://git.kernel.org/linus/e58337100721f3cc0c7424a18730e4f39844934f | https://git.kernel.org/linus/2298e804e96eb3635c39519c8287befd92460303 | ubuntu |
</details>
二、漏洞分析结构反馈
影响性分析说明：
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn't removethe fence from the pending list, and thus doesn't require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it's been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it's held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.The Linux kernel CVE team has assigned CVE-2024-43863 to this issue.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
In the Linux kernel, the following vulnerability has been resolved:drm/vmwgfx: Fix a deadlock in dma buf fence pollingIntroduce a version of the fence ops that on release doesn t removethe fence from the pending list, and thus doesn t require a lock tofix poll->fence wait->fence unref deadlocks.vmwgfx overwrites the wait callback to iterate over the list of allfences and update their status, to do that it holds a lock to preventthe list modifcations from other threads. The fence destroy callbackboth deletes the fence and removes it from the list of pendingfences, for which it holds a lock.dma buf polling cb unrefs a fence after it s been signaled: so the pollcalls the wait, which signals the fences, which are being destroyed.The destruction tries to acquire the lock on the pending fences listwhich it can never get because it s held by the wait from which itwas called.Old bug, but not a lot of userspace apps were using dma-buf pollinginterfaces. Fix those, in particular this fixes KDE stalls/deadlock.
