The arm64 fork-then-swap-out-then-swap-in failures: a program source for exploring them

[Mark Millard]

On 2017-Apr-18, at 11:16 AM, Jia-Shiun Li <jiashiun at gmail.com> wrote:

> just noticed that rpi3 wiki page still has outdated issue info regarding
> this (the jemalloc error). Anyone to help update it?
> 
> https://wiki.freebsd.org/arm64/rpi3

stable/11 is still in process for the fork handling fixes.

One of the 2 fixes to fork behavior has just been MFC'd:
-r313772 from head is now -r317147 in stable/11 . So
interrupts will no longer trash the sp_el0 register.

There is still -r316679 from head to go so that
Copy-On-Write would work for fork. (The defect
may be more general than just being for fork.)

> -Jia-Shiun
> 
> On Mon, Apr 10, 2017 at 5:51 PM, Mark Millard <markmi at dsl-only.net> wrote:
> 
>> buildworld buildkernel completed non-stop for the first time
>> on a BPI-M3 board.
>> 
>> Looks good for a check-in to svn to me (head and stable/11).
>> 
>> This combined with 2017-Feb-15's -r313772's fix to the fork
>> trampline code's updating of sp_el0 makes arm64 far more stable
>> for my purposes.
>> 
>> -r313772 was never MFC'd to stable/11. In my view it should be.



===
Mark Millard
markmi at dsl-only.net