|
|
|
|
|
|
|
|
|
|
xen-ia64-devel
Re: [Xen-ia64-devel] [PATCH] [RFC] [TAKE3] P2M/VP (incomplete) patches
Hi.
On Fri, Mar 24, 2006 at 11:11:38PM +0800, Tian, Kevin wrote:
> A quality writing and good work by far. Due to memory model as the
> most critical/basic component, your work is actually extended to cover
> many areas as the issues posted below. :-) Maybe you have to make a
> priority list, and see whether some core components can be split into
> self-contained parts with major cleanup efforts paid for them first.
Here is the categorized and prioritized list of issues.
These priority is for short/middle term(1-3 months).
Although others might think differently with these priority,
I hope this list helps.
* Issues directly related to the P2M/VP patches.
The following issues are directly related the P2M/VP patches.
Perhaps performance tuning/benchmarking can be done in parallel.
- High discussion on grant table API clean up/performance tuning
- High grant table API clean up
This must be after the discussion.
- High ballon driver
XENMEM_populate_physmap, XENMEM_decrease_reservation,
XENMEM_increase_reservation
- High SMP
Currently nosmp option to xen/IA64 is needed.
This will be addressed after grant table API clean up.
This must be before grant table performance tuning
- High performance tuning related to grant table.
- micro benchmark?
Is there already suitable benchmarks or should it be developed?
- High performance tuning unrelated to grant table.
- Med grant table mapped page reference count and domain destruction
This should be after grant table API clean up and performance tuning.
If a domain is destroyed when it maps a foreign domain's page
via grant table, its page reference count will be leaked.
Page freeing code is needed.
- Low xen_start_info->{console_mfn, store_mfn}
These are defined as machine frame number. However on Xen/IA64 with
P2M/VP model, these should be pseudo physical frame number.
So they should be gmfn instead of mfn.
- Low IOCTL_PRIVCMD_MMAP, IOCTL_PRIVCMD_MMAPBATCH
re-implement xen/IA64 implementation using PageForeign() and
its families.
- Low blktap
* Issues which can be worked independently of P2M/VP model.
The following issues are unrelated the P2M/VP patches or
depends on the stabilized part of P2M/VP patches
so that they can be worked independently of the P2M/VP model patches.
** Platform support
- Low HP ZX1
High for HP(?)
Some modification is necessary for
linux/arch/ia64/hp/common/sba_iommu.c.
- Low SGI Altix
High for SGI(?)
There are many custom drivers under linux/arch/ia64/sn directory.
They also need modifications.
- Low ski simulator
simscsi, simeth are broken. dom0 can't boot.
** Testing
- Low AGP
AGP should work, but I haven't tested it.
- Low virtualized driver
Only vbd and vnif are tested.
Other virtual devices except balloon and blktap should work.
- Low transparent para-virtualization
Some of xen/IA64 developers value transparent para-virtualization.
"if (running_on_xen) { }" is be used, but it isn't tested.
Or is it worthwhile to define a kind of switch which is determined at
boot up time ?
- Low Stability test, benchmark
This might be too early to begin.
** Bug fixes and misc
- High copy_to_guest(), copy_from_guest()
They are broken.
Their copy may success or may result in EFAULT depending on tlb cache
state.
Fortunately xen/PPC port already solved similar problems.
Discussion is needed.
- Low panic_domain()
This function is called when a domain behaves a way xen can't handle
well.
A domain should be stopped, but xen should continue to run.
But the current implementation results in BUG() in xen or drops to
debugger so that xen itself stops.
- Low SUBARCH
- Low dltb miss handler optimization
- Low itlb miss handler fix
fix dom0 startup environment to remove alt itlb miss which occurs
at dom0 boot time.
* My plan
high priority
- discussion on grant table API clean up / performance tuning
- getting balloon driver to work
- grant table API clean up
- SMP
- performance tuning related to grant table
- performance tuning unrelated to grant table.
- stability/testing/benchmark
- merge effort
low priority
--
yamahata
_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel
|
|
|
|
|