linux-nvdimm.lists.01.org archive mirror
 help / color / mirror / Atom feed
From: "Verma, Vishal L" <vishal.l.verma@intel.com>
To: "aneesh.kumar@linux.ibm.com" <aneesh.kumar@linux.ibm.com>,
	"Williams, Dan J" <dan.j.williams@intel.com>
Cc: "linux-nvdimm@lists.01.org" <linux-nvdimm@lists.01.org>
Subject: Re: [PATCH] ndctl: Use the same align value as original namespace on reconfigure
Date: Fri, 18 Oct 2019 17:18:34 +0000	[thread overview]
Message-ID: <4fa50f9c397533f9a615d59af8dd268106a7425e.camel@intel.com> (raw)
In-Reply-To: <87d0eufj03.fsf@linux.ibm.com>


On Fri, 2019-10-18 at 15:55 +0530, Aneesh Kumar K.V wrote:
> Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> writes:
> 
> > "Verma, Vishal L" <vishal.l.verma@intel.com> writes:
> > 
> > > On Wed, 2019-08-07 at 10:14 +0530, Aneesh Kumar K.V wrote:
> > > > When using reconfigure command to add a `name` to the namespace we end
> > > > up updating the align attribute. Avoid this by using the value from
> > > > the original namespace. Do this only if we are keeping the namespace mode
> > > > same.
> > > > 
> > > > Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
> > > > ---
> > > >  ndctl/namespace.c | 16 ++++++++++++++++
> > > >  1 file changed, 16 insertions(+)
> > > 
> > > Hi Aneesh,
> > > 
> > > A few comments below:
> > > 
> > > > diff --git a/ndctl/namespace.c b/ndctl/namespace.c
> > > > index 1f212a2b3a9b..24e51bb35ae1 100644
> > > > --- a/ndctl/namespace.c
> > > > +++ b/ndctl/namespace.c
> > > > @@ -596,6 +596,22 @@ static int validate_namespace_options(struct ndctl_region *region,
> > > >  			return -ENXIO;
> > > >  		}
> > > >  	} else {
> > > > +
> > > > +		/*
> > > > +		 * If we are tryint to reconfigure with the same namespace mode
> > > 
> > >                              ^trying
> > > 
> > > > +		 * Use the align details from the origin namespace. Otherwise
> > > 
> > > s/origin/original/
> > > 
> > > > +		 * pick the align details from seed namespace
> > > > +		 */
> > > > +		if (ndns && p->mode == ndctl_namespace_get_mode(ndns)) {
> > > 
> > > Do we need to depend on the mode here?
> > > 
> > > I'm thinking it should be sufficient to do:
> > >   1. Check We're in 'reconfigure'
> > >   2. Check param.align was not supplied
> > >   3. Get alignment from the pfn/dax personality, and just use that.
> > > 
> > > Does this make sense (Maybe I'm missing something).
> > 
> > We want to use the align value from the seed when we are trying
> > to reconfigure a namespace with a different mode. ie, if we are moving a
> > fsdax namespace with align value 64K to a devdax, IMHO we should pick 16M
> > as alignment for devdax.
> > 
> > > > +			struct ndctl_pfn *ns_pfn = ndctl_namespace_get_pfn(ndns);
> > > > +			struct ndctl_dax *ns_dax = ndctl_namespace_get_dax(ndns);
> > > > +			if (ns_pfn)
> > > > +				p->align = ndctl_pfn_get_align(ns_pfn);
> > > > +			else if (ns_dax)
> > > > +				p->align = ndctl_dax_get_align(ns_dax);
> > > > +			else
> > > > +				p->align = sysconf(_SC_PAGE_SIZE);
> > > 
> > > Do we need the page size fallback here - there are other checks after
> > > this point that also do a similar fallback, do they not catch the
> > > default case?
> > 
> > I did that to simplify the code with that `else if`  
> > 
> > > > +		} else
> > > >  		/*
> > > >  		 * Use the seed namespace alignment as the default if we need
> > > >  		 * one. If we don't then use PAGE_SIZE so the size_align
> 
> Any update on this.?

Yes - I've applied it to the pending branch, and it will be in v67.


_______________________________________________
Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
To unsubscribe send an email to linux-nvdimm-leave@lists.01.org

      reply	other threads:[~2019-10-18 17:18 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-07  4:44 [PATCH] ndctl: Use the same align value as original namespace on reconfigure Aneesh Kumar K.V
2019-08-27  8:39 ` Aneesh Kumar K.V
2019-08-27 18:19 ` Verma, Vishal L
2019-08-28  9:43   ` Aneesh Kumar K.V
2019-10-18 10:25     ` Aneesh Kumar K.V
2019-10-18 17:18       ` Verma, Vishal L [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4fa50f9c397533f9a615d59af8dd268106a7425e.camel@intel.com \
    --to=vishal.l.verma@intel.com \
    --cc=aneesh.kumar@linux.ibm.com \
    --cc=dan.j.williams@intel.com \
    --cc=linux-nvdimm@lists.01.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).