blob: 54b901698f02e9341d0bd176cf86a23056d1b02f (
plain) (
blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
<think>
Alright, I need to classify this bug report for QEMU. Let's go through it step by step.
The problem is about CPU boot failures on big.LITTLE CPUs with varying cache sizes. The affected SoC is RK3588, which has different core clusters (A55 and A76) with differing L2 caches.
In the bug description, QEMU uses `KVM_SET_ONE_REG` to set `CCSIDR`, but if the host moves QEMU between cores, the value changes, leading to an error. The user tried a workaround by modifying a condition in `write_list_to_kvmstate`, which temporarily fixes it but isn't a proper solution.
The issue revolves around how QEMU handles `CCSIDR` across different CPUs with varying cache sizes. It seems related to KVM's interaction with the host kernel, specifically when dealing with CPU-specific registers like CCSIDR.
Looking at the categories provided: hypervisor is relevant here because KVM acts as a hypervisor for virtualization. The problem occurs during CPU initialization in the guest, which ties into how QEMU (the emulator/hypervisor) interacts with the host's KVM.
Other categories like device or network don't fit because the issue isn't hardware devices or networking but rather CPU and cache handling within the hypervisor environment.
So, the appropriate category is "hypervisor" as it pertains to KVM's role in managing guest CPUs and their specific registers.
</think>
hypervisor
|