Linux-Fsdevel Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: Miklos Szeredi <miklos@szeredi.hu>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-nvdimm <linux-nvdimm@lists.01.org>,
	virtio-fs-list <virtio-fs@redhat.com>,
	Dan Williams <dan.j.williams@intel.com>
Subject: Re: [PATCH v3 11/18] fuse: implement FUSE_INIT map_alignment field
Date: Wed, 26 Aug 2020 15:53:11 -0400	[thread overview]
Message-ID: <20200826195311.GB1043442@redhat.com> (raw)
In-Reply-To: <CAJfpegvqZUXsvbWg8K-xosNR+RVwRm2KH+S9mKs6n6Sv65s+Qg@mail.gmail.com>

On Wed, Aug 26, 2020 at 09:26:29PM +0200, Miklos Szeredi wrote:
> On Wed, Aug 26, 2020 at 9:17 PM Dr. David Alan Gilbert
> <dgilbert@redhat.com> wrote:
> 
> > Agreed, because there's not much that the server can do about it if the
> > client would like a smaller granularity - the servers granularity might
> > be dictated by it's mmap/pagesize/filesystem.  If the client wants a
> > larger granularity that's it's choice when it sends the setupmapping
> > calls.
> 
> What bothers me is that the server now comes with the built in 2MiB
> granularity (obviously much larger than actually needed).
> 
> What if at some point we'd want to reduce that somewhat in the client?
>   Yeah, we can't.   Maybe this is not a kernel problem after all, the
> proper thing would be to fix the server to actually send something
> meaningful.

Hi Miklos,

Current implementation of virtiofsd reports this map alignment as
PAGE_SIZE.

        /* This constraint comes from mmap(2) and munmap(2) */
        outarg.map_alignment = ffsl(sysconf(_SC_PAGE_SIZE)) - 1;

Which should be 4K on x86. 

And that means if client wants it can drop to dax mapping size as
small as 4K and still meeting alignment constratints. Just that by
default we have chosen 2MB as of now fearing there might be too
many small mmap() calls on host and we will hit various limits.

Thanks
Vivek


  reply	other threads:[~2020-08-26 19:53 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-19 22:19 [PATCH v3 00/18] virtiofs: Add DAX support Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 01/18] dax: Modify bdev_dax_pgoff() to handle NULL bdev Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 02/18] dax: Create a range version of dax_layout_busy_page() Vivek Goyal
2020-08-20 12:58   ` Jan Kara
2020-08-20 14:29     ` Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 03/18] virtio: Add get_shm_region method Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 04/18] virtio: Implement get_shm_region for PCI transport Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 05/18] virtio: Implement get_shm_region for MMIO transport Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 06/18] virtiofs: Provide a helper function for virtqueue initialization Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 07/18] fuse: Get rid of no_mount_options Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 08/18] virtio_fs, dax: Set up virtio_fs dax_device Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 09/18] fuse,virtiofs: Add a mount option to enable dax Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 10/18] fuse,virtiofs: Keep a list of free dax memory ranges Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 11/18] fuse: implement FUSE_INIT map_alignment field Vivek Goyal
2020-08-26 14:06   ` Miklos Szeredi
2020-08-26 15:51     ` Vivek Goyal
2020-08-26 17:34       ` Stefan Hajnoczi
2020-08-26 19:17         ` Dr. David Alan Gilbert
2020-08-26 19:26           ` Miklos Szeredi
2020-08-26 19:53             ` Vivek Goyal [this message]
2020-08-19 22:19 ` [PATCH v3 12/18] fuse: Introduce setupmapping/removemapping commands Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 13/18] fuse, dax: Implement dax read/write operations Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 14/18] fuse,dax: add DAX mmap support Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 15/18] fuse,virtiofs: Define dax address space operations Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 16/18] fuse, dax: Serialize truncate/punch_hole and dax fault path Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 17/18] fuse,virtiofs: Maintain a list of busy elements Vivek Goyal
2020-08-19 22:19 ` [PATCH v3 18/18] fuse,virtiofs: Add logic to free up a memory range Vivek Goyal
2020-08-28 14:26 ` [PATCH v3 00/18] virtiofs: Add DAX support Miklos Szeredi
2020-08-28 14:39   ` Vivek Goyal

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=20200826195311.GB1043442@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=dan.j.williams@intel.com \
    --cc=dgilbert@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvdimm@lists.01.org \
    --cc=miklos@szeredi.hu \
    --cc=stefanha@redhat.com \
    --cc=virtio-fs@redhat.com \
    /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).