Re: [Minios-devel] Test minios for the arm64

[Huang Shijie <shijie.huang@xxxxxxx>]

Hi Julien,

On Tue, Jun 13, 2017 at 09:32:14AM +0000, Huang Shijie wrote:
> On Fri, Jun 09, 2017 at 11:57:33AM +0100, Julien Grall wrote:
> > (CC correct e-mail for Stefano.)
> > 
> > On 09/06/17 10:52, Huang Shijie wrote:
> > >Hi all,
> > 
> > Hi Shijie,
> > 
> > >   I tested the Volodymyr's git tree with arm64 platform (Softiron)
> > >   The mini-os hang when it tried to enable the MMU.
> > >
> > >   The following is the log:
> > >  --------------------------------------------------------
> > >   (XEN) Freed 288kB init memory.
> > >   - Mini-OS booting -
> > >   - Setup CPU -
> > >   - Setup booting pagetable -
> > >   - Setup ID mapping -
> > >   *** Sync exception at PC = ffffffc000001048 ***
> > >   Thread state:
> > >   X0  = 0x0000000000000000 X1  = 0x0000000034d5d905
> > >   X2  = 0xffffffc000000800 X3  = 0x0000000000000100
> > >   X4  = 0x000000800801f000 X5  = 0x0000008008022000
> > >   X6  = 0x0000008008000000 X7  = 0x0000008008020003
> > >   X8  = 0x0000000000000000 X9  = 0x0000000008008020
> > >   X10 = 0x0000000000000000 X11 = 0x0000000000000000
> > >   X12 = 0x0000000000000000 X13 = 0x0000000000000000
> > >   X14 = 0x0000000000000000 X15 = 0x0000000000000000
> > >   X16 = 0x00000000deadbeef X17 = 0x0000000000000000
> > >   X18 = 0x0000008000000711 X19 = 0xffffffc000000800
> > >   X20 = 0x000000800fe00000 X21 = 0xffffffc000400020
> > >   X22 = 0xffffffc000000fc4 X23 = 0x00000000600003c5
> > >   X24 = 0x0000000000000000 X25 = 0x0000000000000000
> > >   X26 = 0x0000000000000000 X27 = 0x0000000000000000
> > >   X28 = 0x0000000000000000 X29 = 0xffffffc000000800
> > >   X30 (lr) = 0x00000080080001ac
> > >   sp  = 0xffffffc0003fff10
> > >   esr_el1 = 96000046
> > 
> > This is a data abort. I would when trying to access...
> > 
> > >   far_el1 = ffffffc000400000
> > 
> > this address. Have would recommend to look what is the instruction (and
> > function) which generates the data abort.
> This address (far_el1 = ffffffc000400000) is used for stack.
> 
> I guess that an exception nests in another exception.
> 
> The x22(0xffffffc000000fc4) (which is the older ELR_EL1) is the first 
> exception,
> 
> In the trap_entry() code, the first "push" triggers the second exception.

I find the code does not implement the page table mapping for the stack memory.
So when the mmu is enabled, the minios will hang.


Thanks
Huang Shijie

_______________________________________________
Minios-devel mailing list
Minios-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/cgi-bin/mailman/listinfo/minios-devel