Re: [Xen-devel] Possible issue with x86_emulate when writing results back to memory

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 09.01.14 at 17:07, Simon Graham <simon.graham@xxxxxxxxxx> wrote:
>> > The attached patch is for discussion purposes only - if it is deemed
>> > acceptable I'll resubmit a proper patch request against unstable.
>> 
>> I'd rather not add limited scope special casing like that, but instead
>> make the copying much more like real hardware (i.e. not just deal
>> with the 16- and 32-bit cases, and especially not rely on memcpy()
>> using 64-bit reads/writes when it can). IOW - don't use memcpy()
>> here at all (and have a single routine doing The Right Thing (tm)
>> rather than having two clones now, and perhaps more later on -
>> I'd in particular think that the read side in shadow code would also
>> need a similar adjustment).
> 
> My concern was that memcpy is (I assume!) highly optimized - it certainly 
> should be if it isn't and I would worry that a change to make it atomic for 
> the purposes of instruction emulation would result in an across the board 
> perf hit when in most cases it isn't necessary that it be atomic.

And I didn't mean to fiddle with memcpy(), but rather create a
specialized copying function just for the use in the context of
emulation.

Jan


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