Re: KVM "fake DAX" flushing interface - discussion

2017-11-22 Thread Xiao Guangrong
On 11/22/2017 02:19 AM, Rik van Riel wrote: We can go with the "best" interface for what could be a relatively slow flush (fsync on a file on ssd/disk on the host), which requires that the flushing task wait on completion asynchronously. I'd like to clarify the interface of "wait on

Re: KVM "fake DAX" flushing interface - discussion

2017-11-03 Thread Xiao Guangrong
On 11/03/2017 12:30 AM, Dan Williams wrote: On Thu, Nov 2, 2017 at 1:50 AM, Xiao Guangrong <xiaoguangrong.e...@gmail.com> wrote: [..] Yes, the GUID will specifically identify this range as "Virtio Shared Memory" (or whatever name survives after a bikeshed debate). Th

Re: KVM "fake DAX" flushing interface - discussion

2017-11-02 Thread Xiao Guangrong
On 11/01/2017 11:20 PM, Dan Williams wrote: On 11/01/2017 12:25 PM, Dan Williams wrote: [..] It's not persistent memory if it requires a hypercall to make it persistent. Unless memory writes can be made durable purely with cpu instructions it's dangerous for it to be treated as a PMEM range.

Re: KVM "fake DAX" flushing interface - discussion

2017-11-01 Thread Xiao Guangrong
On 11/01/2017 12:25 PM, Dan Williams wrote: On Tue, Oct 31, 2017 at 8:43 PM, Xiao Guangrong <xiaoguangrong.e...@gmail.com> wrote: On 10/31/2017 10:20 PM, Dan Williams wrote: On Tue, Oct 31, 2017 at 12:13 AM, Xiao Guangrong <xiaoguangrong.e...@gmail.com> wrote: On 07/27/2

Re: KVM "fake DAX" flushing interface - discussion

2017-10-31 Thread Xiao Guangrong
On 10/31/2017 10:20 PM, Dan Williams wrote: On Tue, Oct 31, 2017 at 12:13 AM, Xiao Guangrong <xiaoguangrong.e...@gmail.com> wrote: On 07/27/2017 08:54 AM, Dan Williams wrote: At that point, would it make sense to expose these special virtio-pmem areas to the guest in a slightly dif

Re: KVM "fake DAX" flushing interface - discussion

2017-10-31 Thread Xiao Guangrong
On 07/27/2017 08:54 AM, Dan Williams wrote: At that point, would it make sense to expose these special virtio-pmem areas to the guest in a slightly different way, so the regions that need virtio flushing are not bound by the regular driver, and the regular driver can continue to work for

Re: DAX mapping detection (was: Re: [PATCH] Fix region lost in /proc/self/smaps)

2016-09-12 Thread Xiao Guangrong
On 09/12/2016 11:44 AM, Rudoff, Andy wrote: Whether msync/fsync can make data persistent depends on ADR feature on memory controller, if it exists everything works well, otherwise, we need to have another interface that is why 'Flush hint table' in ACPI comes in. 'Flush hint table' is

Re: [PATCH] acpi, nfit: fix acpi_check_dsm() vs zero functions implemented

2016-07-20 Thread Xiao Guangrong
On 07/21/2016 06:49 AM, Dan Williams wrote: On Tue, Jul 19, 2016 at 3:53 PM, Dan Williams wrote: On Tue, Jul 19, 2016 at 3:46 PM, Jerry Hoemann wrote: On Tue, Jul 19, 2016 at 01:01:16PM -0700, Dan Williams wrote: On Tue, Jul 19, 2016 at

Re: [PATCH v2 1/2] nfit: fix _FIT evaluation memory leak

2016-07-18 Thread Xiao Guangrong
..@intel.com> Reported-by: Xiao Guangrong <guangrong.x...@intel.com> Reported-by: Haozhong Zhang <haozhong.zh...@intel.com> Signed-off-by: Dan Williams <dan.j.willi...@intel.com> --- Change since v1: * Fix unitialized use of 'rc' (Haozhong) * Clarify that their is no use-after

Re: [PATCH 1/2] nfit: fix _FIT evaluation memory leak

2016-07-14 Thread Xiao Guangrong
On 07/15/2016 11:28 AM, Dan Williams wrote: acpi_evaluate_object() allocates memory. Free the buffer allocated during acpi_nfit_add(). Dan, thanks for your fix. Another one is the use-after-free issue in acpi_nfit_notify(): /* Evaluate _FIT */ status =