On Sat, Aug 27, 2016 at 5:21 AM, Pavel Machek <[email protected]> wrote:
>
> Hi!
>
>> Expose a new system call allowing each thread to register one userspace
>> memory area to be used as an ABI between kernel and user-space for two
>> purposes: user-space restartable sequences and quick access to read the
>> current CPU number value from user-space.
>>
>> * Restartable sequences (per-cpu atomics)
>>
>> Restartables sequences allow user-space to perform update operations on
>> per-cpu data without requiring heavy-weight atomic operations.
>>
>> The restartable critical sections (percpu atomics) work has been started
>> by Paul Turner and Andrew Hunter. It lets the kernel handle restart of
>> critical sections. [1] [2] The re-implementation proposed here brings a
>> few simplifications to the ABI which facilitates porting to other
>> architectures and speeds up the user-space fast path. A locking-based
>> fall-back, purely implemented in user-space, is proposed here to deal
>> with debugger single-stepping. This fallback interacts with rseq_start()
>> and rseq_finish(), which force retries in response to concurrent
>> lock-based activity.
>
> Hmm. Purely software fallback needed for singlestepping... Looks like this is malware
> writer's dream come true...
>
> Also if you ever get bug in the restartable code, debugger will be useless to debug it...
> unless new abilities are added to debuggers to manually schedule threads on CPUs.
>
> Is this good idea?
We've had some off-list discussion.
I have a revised version which incoprorates some of Mattheiu's
improvements, but avoids this requirement nearly ready for posting.
- Paul