Alpine Linux Issue

[evonox]

Hi all, I ran into this error with the build of Alpine distro. I use the latest version of the emulator and I have built Alpine using the Docker.

What am I doing wrong?

[SuperMaxusa]

About NMI selftest errors: probably https://github.com/copy/v86/issues/797 About IRQ errors: can you try adding irqpoll to cmdline?

--- a/alpine.html
+++ b/alpine.html
@@ -18,7 +18,7 @@ window.onload = function()
         },
         autostart: true,
         bzimage_initrd_from_filesystem: true,
-        cmdline: "rw root=host9p rootfstype=9p rootflags=trans=virtio,cache=loose modules=virtio_pci tsc=reliable",
+        cmdline: "rw root=host9p rootfstype=9p rootflags=trans=virtio,cache=loose modules=virtio_pci tsc=reliable irqpoll",
         //initial_state: { url: "../images/alpine-state.bin" },
     });
 };

[coderofsalvation]

I've tried this fix, it gives this (same?) output:

[   62.464666]  ? __irq_disable+0xa0/0xa0                                       
[   62.464666]  ? handle_level_irq+0x7c/0x160                                   
[   62.464666]  ? __handle_irq+0x9e/0x100                                       
[   62.464666]  </IRQ>                                                          
[   62.464666]  ? __common_interrupt+0x36/0xa0                                  
[   62.464666]  ? common_interrupt+0x2a/0x50                                    
[   62.464666]  ? asm_common_interrupt+0x102/0x140                              
[   62.464666]  ? __SCT__tp_func_ipi_exit+0x8/0x8                               
[   62.464666]  ? __SCT__tp_func_ipi_exit+0x8/0x8                               
[   62.464666]  ? delay_tsc+0xbd/0xe0                                           
[   62.464666]  __const_udelay+0x31/0x40                                        
[   62.464666]  test_nmi_ipi.constprop.0+0x98/0xf0                              
[   62.464666]  nmi_selftest+0x140/0x3b0                                        
[   62.464666]  native_smp_cpus_done+0x157/0x270                                
[   62.464666]  smp_init+0x6e/0xa0                                              
[   62.464666]  kernel_init_freeable+0x143/0x450                                
[   62.464666]  ? rest_init+0xc0/0xc0                                           
[   62.464666]  ? calculate_sigpending+0x2f/0x40                                
[   62.464666]  ? rest_init+0xc0/0xc0                                           
[   62.464666]  kernel_init+0x17/0x1a0                                          
[   62.464666]  ret_from_fork+0x38/0x60                                         
[   62.464666]  ? rest_init+0xc0/0xc0                                           
[   62.464666]  ret_from_fork_asm+0x12/0x20                                     
[   62.464666]  entry_INT80_32+0xf0/0xf5   

iirc correctly it had something to do with later Alpine kernels being built with a certain different default. (TBH I've googled this a while ago (when I ran into it) and read somewhere (can't find the link) )

[SuperMaxusa]

it gives this (same?) output

If I correctly understand, this message can be repeated several times in log. On Node.JS (for example, run build-state.js) does this also happen?

[evonox]

What might be the causes that the kernel cannot cannot find host9p device? Should the 9p FS drivers be statically linked to kernel or can they be dynamically loaded?

[SuperMaxusa]

What might be the causes that the kernel cannot cannot find host9p device?

The 9p(net): ... lines are appeared in your log? Also check these kernel parameters: rootfstype=9p rootflags=trans=virtio

[evonox]

Thank you. I was bulding the old distro Ubuntu 18.04 which kernel did not have support for 9P. I have built the latest kernel available that has it turned on by default and it works.

[copy]

This is the same as https://github.com/copy/v86/issues/797. There's a fix in the wip branch.

[copy]

This should be fixed now.