Re: [PATCH 1/2] perf: Add persistent event facilities

[Peter Zijlstra]

On Fri, 2012-05-18 at 14:55 +0200, Borislav Petkov wrote:
> It is an interesting question how we handle that RO/RW thing?

RO buffers overwrite, RW buffers loose events.

We could allow both for persistent under the constraints:

 VM_SHARED - must imply !VM_WRITE since multiple people writing to the
             control page will get 'interesting' very fast.

 !VM_SHARED - must fail mmap() of the debugfs file if there's already
              an existing user.

Note that even RW buffers only allows writes to the control page, a
write to the actual data itself will generate a fault.
--
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/