Hi,
Just a quick question: What is the status of the two projects in
$subject? Havn't heared anything for a while for both ...
cheers,
Gerd
--
http://kraxel.fedorapeople.org/xenner/
___
Virtualization mailing list
Xen has a pte update function which will update a pte while preserving
its accessed and dirty bits. This means that pte_rmw_start() can be
implemented as a simple read of the pte value. The hardware may
update the pte in the meantime, but pte_rmw_commit() updates it while
preserving any changes
Some Xen hypercalls accept an array of operations to work on. In
general this is because its more efficient for the hypercall to the
work all at once rather than as separate hypercalls (even batched as a
multicall).
This patch adds a mechanism (xen_mc_extend_args()) to allocate more
argument
Hi all,
This little series adds a new transaction-like abstraction for doing
RMW updates to a pte, hooks it into paravirt_ops, and then makes use
of it in Xen.
The basic problem is that mprotect is very slow under Xen (up to 50x
slower than native), primarily because of the
ptent =
This patch adds an API for doing read-modify-write updates to a pte
which may race against hardware updates to the pte. After reading the
pte, the hardware may asynchonously set the accessed or dirty bits on
a pte, which would be lost when writing back the modified pte value.
The existing
Zachary Amsden wrote:
I'm a bit skeptical you can get such a semantic to work without a very
heavyweight method in the hypervisor. How do you guarantee no other CPU
is fizzling the A/D bits in the page table (it can be done by hardware
with direct page tables), unless you use some kind of
Linus Torvalds wrote:
On Fri, 23 May 2008, Jeremy Fitzhardinge wrote:
This series adds the pte_rmw_start() and pte_rmw_commit() operations,
which change this sequence to:
ptent = pte_rmw_start(mm, addr, pte);
ptent = pte_modify(ptent, newprot);
/* ... */
On Fri, 2008-05-23 at 21:32 +0100, Jeremy Fitzhardinge wrote:
Zachary Amsden wrote:
I'm a bit skeptical you can get such a semantic to work without a very
heavyweight method in the hypervisor. How do you guarantee no other CPU
is fizzling the A/D bits in the page table (it can be done by