All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: Alex Williamson <alex.williamson@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>,
	linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
	peterx@redhat.com
Subject: Re: [PATCH 3/7] vfio/pci: Use vfio_device_unmap_mapping_range()
Date: Wed, 11 Aug 2021 08:57:14 -0300	[thread overview]
Message-ID: <20210811115714.GB7008@nvidia.com> (raw)
In-Reply-To: <20210810155058.4199a86b.alex.williamson@redhat.com>

On Tue, Aug 10, 2021 at 03:50:58PM -0600, Alex Williamson wrote:
> On Tue, 10 Aug 2021 13:55:00 +0100
> Christoph Hellwig <hch@infradead.org> wrote:
> 
> > On Tue, Aug 10, 2021 at 08:57:22AM -0300, Jason Gunthorpe wrote:
> > > I'm not sure there is a real performance win to chase here? Doesn't
> > > this only protect mmap against reset? The mmap isn't performance
> > > sensitive, right?
> > > 
> > > If this really needs extra optimization adding a rwsem to the devset
> > > and using that across the whole set would surely be sufficient.  
> > 
> > Every mmio read or write takes memory_lock.
> 
> Exactly.  Ideally we're not using that path often, but I don't think
> that's a good excuse to introduce memory access serialization, or even
> dependencies between devices.  Thanks,

But a cross device rwsem seems OK to me?? It won't contend unless we
are trying to reset and the upgrade to a percpu rwsem to optimize
the atomic doesn't seem warranted since this path already has a vmexit
and a syscall in it?

Jason

  reply	other threads:[~2021-08-11 11:57 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-05 17:06 [PATCH 0/7] vfio: device fd address space and vfio-pci mmap invalidation cleanup Alex Williamson
2021-08-05 17:07 ` [PATCH 1/7] vfio: Create vfio_fs_type with inode per device Alex Williamson
2021-08-10  8:43   ` Christoph Hellwig
2021-08-10 14:52     ` Alex Williamson
2021-08-10 14:57       ` Christoph Hellwig
2021-08-10 18:49         ` Peter Xu
2021-08-10 21:16           ` Alex Williamson
2021-08-10 22:18             ` Peter Xu
2021-08-05 17:07 ` [PATCH 2/7] vfio: Export unmap_mapping_range() wrapper Alex Williamson
2021-08-10  8:45   ` Christoph Hellwig
2021-08-10 18:56   ` Peter Xu
2021-08-05 17:07 ` [PATCH 3/7] vfio/pci: Use vfio_device_unmap_mapping_range() Alex Williamson
2021-08-06  1:04   ` Jason Gunthorpe
2021-08-06 20:17     ` Alex Williamson
2021-08-10  8:51       ` Christoph Hellwig
2021-08-10 11:57         ` Jason Gunthorpe
2021-08-10 12:55           ` Christoph Hellwig
2021-08-10 21:50             ` Alex Williamson
2021-08-11 11:57               ` Jason Gunthorpe [this message]
2021-08-10  8:53   ` Christoph Hellwig
2021-08-10 19:02     ` Peter Xu
2021-08-10 20:51       ` Alex Williamson
2021-08-10 18:48   ` Peter Xu
2021-08-10 19:59     ` Alex Williamson
2021-08-10 20:20       ` Peter Xu
2021-08-05 17:07 ` [PATCH 4/7] vfio,vfio-pci: Add vma to pfn callback Alex Williamson
2021-08-06  1:01   ` Jason Gunthorpe
2021-08-10  9:00     ` Christoph Hellwig
2021-08-10  9:00   ` Christoph Hellwig
2021-08-05 17:08 ` [PATCH 5/7] mm/interval_tree.c: Export vma interval tree iterators Alex Williamson
2021-08-05 17:08 ` [PATCH 6/7] vfio: Add vfio_device_io_remap_mapping_range() Alex Williamson
2021-08-10  9:04   ` Christoph Hellwig
2021-08-05 17:08 ` [PATCH 7/7] vfio/pci: Remove map-on-fault behavior Alex Williamson
2021-08-10  9:11   ` Christoph Hellwig
2021-08-10 15:04     ` Alex Williamson
2021-08-10 20:54   ` Peter Xu
2021-08-10 21:45     ` Alex Williamson
2021-08-10 22:27       ` Peter Xu

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=20210811115714.GB7008@nvidia.com \
    --to=jgg@nvidia.com \
    --cc=alex.williamson@redhat.com \
    --cc=hch@infradead.org \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterx@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.