|
|
|
|
|
|
|
|
|
|
xen-ia64-devel
Re: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g
On Fri, Mar 31, 2006 at 09:35:23AM +0100, Tristan Gingold wrote:
> Le Vendredi 31 Mars 2006 08:15, Isaku Yamahata a écrit :
> [...]
> > > The IA64_GRANULE_SIZE (16M/64M) area covering current kernel stack
> > > is also mapped by TR. Except for init_task, all kernel stacks are
> > > dynamically allocated from identity mapping area. So if stack of current
> > > running task happens to reside within [PAGE_OFFSET, PAGE_OFFSET
> > > + IA64_GRANULE_SIZE - 1], then that area is mapped by TR. Then once
> > > stack of next task is outside of that range, TR will be purged and
> > > re-inserted upon new area.
> > >
> > > So I think Isaku is only taking [PAGE_OFFSET, PAGE_OFFSET+64M] as
> > > an example to explain the issue, with assumption that area doesn't
> > > contain current kernel stack. :-)
> >
> > Exactly.
> What I still do not understand is why page_fault would be injected to linux.
> For me, it is alt dtlb miss, which always issues a itc for region 7 IIRC.
You are right. Linux should see alt dtlb miss fault.
--
yamahata
_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel
|
<Prev in Thread] |
Current Thread |
[Next in Thread> |
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, (continued)
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Tian, Kevin
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Tian, Kevin
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Xu, Anthony
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Xu, Anthony
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Xu, Anthony
- RE: [Xen-ia64-devel] Re: [PATCH]: ptc.ga for SMP-g, Tian, Kevin
|
|
|
|
|