Re: [Xen-devel] [PATCH 2/2] xen/livepatch: Don't crash on encountering STN_UNDEF relocations

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 13.06.17 at 23:13, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 13/06/17 21:51, Andrew Cooper wrote:
>> A symndx of STN_UNDEF is special, and means a symbol value of 0.
>>
>> There is no real symbol data for it, so avoid tripping over a NULL pointer
>> with "elf->sym[symndx].sym->st_value".
>>
>> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
>> ---
>> CC: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
>> CC: Ross Lagerwall <ross.lagerwall@xxxxxxxxxx>
>> CC: Jan Beulich <JBeulich@xxxxxxxx>
>> CC: Stefano Stabellini <sstabellini@xxxxxxxxxx>
>> CC: Julien Grall <julien.grall@xxxxxxx>
>>
>> Functionally tested on x86, but both arm variants look to suffer from the 
> same
>> issue.  Compile tested on all architectures.
>>
>> TODO: Figure out how my livepatch has a STN_UNDEF relocation...
> 
> On second thoughts, maybe STN_UNDEF symbols should be a hard failure.

Perhaps, but I still don't see where it was coming from in the first
place.

> (XEN) livepatch: live: Applying 1 functions
> (XEN) ----[ Xen-4.10-unstable  x86_64  debug=y   Not tainted ]----
> (XEN) CPU:    3
> (XEN) RIP:    e008:[<0000000000000000>] 0000000000000000
> 
> In this case, the hook function hasn't been wired up correctly, which
> causes apply_payload() to fall over a NULL data->load_funcs[i]();
> 
> As for why the STN_UNDEF symbol, it comes from an assembly hook function
> missing a .type attribute.  Still, Xen shouldn't crash when it
> encounters one.

I don't understand - a missing .type would lead to STT_NOTYPE in
the symbol table, but not a relocation targeting STN_UNDEF.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel