blob: 0d3b104a1c0182925c98c91db73e7e9ac9037feb (
plain) (
blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
|
qemu-system-m68k virt machine doesn't boot Linux kernels using 68020, 68030 and 68060 CPUs
Description of problem:
QEMU doesn't seem to be able to start Linux kernels using a CPU other than a 68040 (which does work fine)
To rule out host issues, the issue is reproductible on Debian Unstable amd64 (with version QEMU emulator version 9.2.0)(Debian 1:9.2.0+ds-5))
To rule out cross-compilation issues, the kernel has been rebuild inside a virt machine (using a 68040 CPU), running Debian Unstable
Each CPU model below gets stuck early before kernel boot during the ABCGHIJK thing. The Kernel doesn't seem to boot and QEMU process eat 100% of a CPU physical core
**68020**
```
qemu-system-m68k -M virt -cpu m68060 -m 1G -nographic -kernel /home/demik/tmp/vmlinux
ABCGH
```
**68030**
```
qemu-system-m68k -M virt -cpu m68060 -m 1G -nographic -kernel /home/demik/tmp/vmlinux
ABC
```
**68060**
```
qemu-system-m68k -M virt -cpu m68060 -m 1G -nographic -kernel /home/demik/tmp/vmlinux
ABC
```
Steps to reproduce:
1. build a kernel with 68020/030/060 support (using virt_defconfig as base)
2. start QEMU with the command line above
Additional information:
68020 is understandable as it may need some sort of 68851 emulation.
Relevant Kernel config Processor configuration:
```
#
# Processor Type
#
CONFIG_M68KCLASSIC=y
# CONFIG_COLDFIRE is not set
CONFIG_M68020=y
CONFIG_M68030=y
CONFIG_M68040=y
CONFIG_M68060=y
```
This may be related to the following issues (but I don't have the skillset to confirm that)
- https://gitlab.com/qemu-project/qemu/-/issues/2091
- https://gitlab.com/qemu-project/qemu/-/issues/2500
|