Re: [RFC PATCH] percpu system call: fast userspace percpu critical sections

[Mathieu Desnoyers]

----- Original Message -----
> On Thu, May 21, 2015 at 7:44 AM, Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx> wrote:
> > Expose a new system call allowing userspace threads to register
> > a TLS area used as an ABI between the kernel and userspace to
> > share information required to create efficient per-cpu critical
> > sections in user-space.
> 
> Not a way in hell.
> 
> You do user accesses from preempt notifiers.
> 
> That's so ludicrously broken that it's not even funny.
> 
> This patch needs to die, and never *ever* be resurrected. Put a stake
> in it, and bury it 6ft deep.

There is no need to fetch your vampire hunting kit just yet. ;-)

As an RFC patch, this should be considered as a prototype, nothing more.
I was mainly expecting feedback on the general approach, and trying to
put some pressure on Paul Turner, so he would post the patches he
talked about back in 2013.

Peter Zijlstra already pointed out that I should move those user accesses
before return to userspace.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/