On 05/10/18 11:48, Wei Liu wrote:
> Got this one-off crash while booting staging (d36b770458) on a skylake
> server. After rebooting it went away.
>
> (XEN) Assertion '(sp == 0) || (peoi[sp-1].vector < vector)' failed at 
> irq.c:1173          
> (XEN) ----[ Xen-4.12-unstable  x86_64  debug=y   Tainted:  C   ]----          
>                
> (XEN) CPU:    5                                                               
>                            
> (XEN) RIP:    e008:[<ffff82d080286921>] do_IRQ+0x496/0x680                    
>             
> (XEN) RFLAGS: 0000000000010046   CONTEXT: hypervisor                          
>             
> (XEN) rax: ffff83085df7a4c0   rbx: ffff83085df81e00   rcx: 0000000000000001   
>             
> (XEN) rdx: 0000000000000021   rsi: 0000000000000021   rdi: 0000000000000001   
>             
> (XEN) rbp: ffff83085df77d98   rsp: ffff83085df77d38   r8:  0000000000000021   
>             
> (XEN) r9:  0000000000000000   r10: 0000000000000000   r11: 0000000000000000   
>  
> (XEN) r12: ffff8308740e4f10   r13: 0000000000000021   r14: ffff83085df81e00   
>   
> (XEN) r15: 000000000000001e   cr0: 0000000080050033   cr4: 00000000003526e0   
>   
> (XEN) cr3: 000000085da9e000   cr2: 00007fc5b6a8cfe8                           
>   
> (XEN) fsb: 0000000000000000   gsb: 0000000000000000   gss: 0000000000000000   
>   
> (XEN) ds: 0000   es: 0000   fs: 0000   gs: 0000   ss: 0000   cs: e008         
>   
> (XEN) Xen code around <ffff82d080286921> (do_IRQ+0x496/0x680):                
>   
> (XEN)  be 00 00 00 7e 93 0f 0b <0f> 0b 0f 0b 0f 0b b8 00 00 00 00 eb 4e 83 bb 
> 1c
> (XEN) Xen stack trace from rsp=ffff83085df77d38:                              
>   
> (XEN)    ffff82d000000000 ffff83085df81e24 0000000000000000 0000001e8037a835  
>   
> (XEN)    ffff82d08037a841 ffff82d08037a835 ffff82d08037a841 0000000000000000  
>      
> (XEN)    0000000000000000 0000000000000000 ffff83085df77fff 0000000000000000  
>      
> (XEN)    00007cf7a2088237 ffff82d08037a8aa 0000000380f0b241 0000000000000008  
>      
> (XEN)    ffff83085df79448 ffff83085df79390 ffff83085df77ec0 0000000380f62e26  
>      
> (XEN)    00000003810cc680 ffff8307de5670a8 00000000001f644f 0000000000000809  
>      
> (XEN)    ffff83085df7a02c 0000000000000000 ffff83085df77fff 00000000000051f3  
>      
> (XEN)    ffff83085df793c0 0000002100000000 ffff82d0802e1684 000000000000e008  
>      
> (XEN)    0000000000000202 ffff83085df77e50 0000000000000000 ffff82d08059bc80  
>      
> (XEN)    00000020ffffffff ffff83085df77fff ffff82d0805a3c80 ffff83085df77eb0  
>                                    
> (XEN)    0000000000000000 0000000000000000 0000033b00000212 ffff82d08059bf00  
>   
> (XEN)    0000000000000005 ffff82d08059bf00 0000000000000005 0000000000000005  
>   
> (XEN)    ffff83085df39000 ffff83085df77ef0 ffff82d0802770b8 ffff830864159000  
>   
> (XEN)    ffff8300791fd000 ffff8300791fb000 ffff830864159000 ffff83085df77db8  
>   
> (XEN)    0000000000000000 0000000000000000 ffff88017dbe3d00 ffff88017dbe3d00  
>   
> (XEN)    0000000000000002 0000000000000002 0000000000000000 0000000000000000  
>        
> (XEN)    0000000148106000 000000005236fe09 ffffffff816fe980 ffff880182a9a7c0  
>  
> (XEN)    ffffffff82049af8 ffff880182a9a7c0 0000000000000082 0000beef0000beef  
>   
> (XEN)    ffffffff816fec52 000000bf0000beef 0000000000000246 ffffc900006dbe98  
>   
> (XEN)    000000000000beef 000000000000beef 000000000000beef 000000000000beef  
>   
> (XEN) Xen call trace:                                                         
>   
> (XEN)    [<ffff82d080286921>] do_IRQ+0x496/0x680                              
>        
> (XEN)    [<ffff82d08037a8aa>] common_interrupt+0x10a/0x120                    
>                
> (XEN)    [<ffff82d0802e1684>] mwait-idle.c#mwait_idle+0x296/0x372             
>    
> (XEN)    [<ffff82d0802770b8>] domain.c#idle_loop+0xb3/0xb5                    
>   
> (XEN)                                                                         
>   
> (XEN)                                                                         
>   
> (XEN) ****************************************                                
>   
> (XEN) Panic on CPU 5:                                                         
>   
> (XEN) Assertion '(sp == 0) || (peoi[sp-1].vector < vector)' failed at 
> irq.c:1173
> (XEN) ****************************************                                
>   
> (XEN)                                                                         
>   
> (XEN) Manual reset required ('noreboot' specified)  
>
> Let me know what else is needed.

We've seen this reported sporadically in the past, and never with enough
information to investigate.  I had one reliable repro of the issue in
the past, which disappeared with a microcode update.

It is always out of mwait, and Xen's logic for which interrupts are
pending now disagrees with hardware.  This means we've seen an interrupt
at a lower priority that one we believe to be pending, which is (to a
first approximation), a violation of LAPIC priority logic.

Perhaps I should insert a tonne of debugging in place of this assertion,
with the hope that the next time we randomly encounter it, we've got
some better idea of what is going on.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Reply via email to