Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Dan Williams
On Thu, Nov 10, 2016 at 8:50 PM, Elliott, Robert (Persistent Memory) wrote: > > >> -Original Message- >> From: Linux-nvdimm [mailto:linux-nvdimm-boun...@lists.01.org] On >> Behalf Of Vishal Verma >> Sent: Thursday, November 10, 2016 3:56 PM >> To: Dan Williams

RE: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Elliott, Robert (Persistent Memory)
> -Original Message- > From: Linux-nvdimm [mailto:linux-nvdimm-boun...@lists.01.org] On > Behalf Of Vishal Verma > Sent: Thursday, November 10, 2016 3:56 PM > To: Dan Williams > Cc: linux-nvdimm@lists.01.org > Subject: Re: [PATCH v2] libnvdimm: check and clear

Re: [PATCH] mm: add ZONE_DEVICE statistics to smaps

2016-11-10 Thread Anshuman Khandual
On 11/11/2016 03:41 AM, Dan Williams wrote: > ZONE_DEVICE pages are mapped into a process via the filesystem-dax and > device-dax mechanisms. There are also proposals to use ZONE_DEVICE > pages for other usages outside of dax. Add statistics to smaps so > applications can debug that they are

Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Vishal Verma
On 11/10, Dan Williams wrote: > On Thu, Nov 10, 2016 at 1:56 PM, Vishal Verma > wrote: > > On 11/10, Dan Williams wrote: > >> On Thu, Nov 10, 2016 at 11:17 AM, Vishal Verma > >> wrote: > >> > On 11/10, Dave Jiang wrote: > >> >> We need to

Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Kani, Toshimitsu
On Thu, 2016-11-10 at 14:56 -0700, Vishal Verma wrote: > On 11/10, Dan Williams wrote:  : > > This feels like over-engineering a still not perfect solution to a > > rare problem.  Outside of atomic-write-and-clear we should just > > keep the code best effort and simple. > > Fair enough :) In that

Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Dan Williams
On Thu, Nov 10, 2016 at 1:56 PM, Vishal Verma wrote: > On 11/10, Dan Williams wrote: >> On Thu, Nov 10, 2016 at 11:17 AM, Vishal Verma >> wrote: >> > On 11/10, Dave Jiang wrote: >> >> We need to clear any poison when we are writing to pmem.

Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Vishal Verma
On 11/10, Dan Williams wrote: > On Thu, Nov 10, 2016 at 11:17 AM, Vishal Verma > wrote: > > On 11/10, Dave Jiang wrote: > >> We need to clear any poison when we are writing to pmem. The granularity > >> will be sector size. If it's less then we can't do anything about

Re: [PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Dan Williams
On Thu, Nov 10, 2016 at 11:17 AM, Vishal Verma wrote: > On 11/10, Dave Jiang wrote: >> We need to clear any poison when we are writing to pmem. The granularity >> will be sector size. If it's less then we can't do anything about it >> barring corruption. >> >>

Re: [patch] nd_blk,nd_pmem,nd_btt: add endio blktrace events

2016-11-10 Thread Christoph Hellwig
On Wed, Nov 09, 2016 at 02:43:58PM -0500, Jeff Moyer wrote: > But on the issue side, we have different trace actions: Q vs. I. On the > completion side, we just have C. You'd end up getting two C events for > each Q, and that may confuse existing utilities (such as blkparse, btt, > iowatcher,

[PATCH v2] libnvdimm: check and clear poison before writing to pmem

2016-11-10 Thread Dave Jiang
We need to clear any poison when we are writing to pmem. The granularity will be sector size. If it's less then we can't do anything about it barring corruption. Signed-off-by: Dave Jiang --- drivers/nvdimm/claim.c | 24 +--- 1 file changed, 21