All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: Alistair Popple <apopple@nvidia.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	will@kernel.org, catalin.marinas@arm.com, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org, robin.murphy@arm.com,
	nicolinc@nvidia.com, linux-arm-kernel@lists.infradead.org,
	kvm@vger.kernel.org, John Hubbard <jhubbard@nvidia.com>,
	zhi.wang.linux@gmail.com, Sean Christopherson <seanjc@google.com>
Subject: Re: [PATCH 2/2] arm64: Notify on pte permission upgrades
Date: Sat, 27 May 2023 21:02:13 -0300	[thread overview]
Message-ID: <ZHKaBQt8623s9+VK@nvidia.com> (raw)
In-Reply-To: <5d8e1f752051173d2d1b5c3e14b54eb3506ed3ef.1684892404.git-series.apopple@nvidia.com>

On Wed, May 24, 2023 at 11:47:29AM +1000, Alistair Popple wrote:
> ARM64 requires TLB invalidates when upgrading pte permission from
> read-only to read-write. However mmu_notifiers assume upgrades do not
> need notifications and none are sent. This causes problems when a
> secondary TLB such as implemented by an ARM SMMU doesn't support
> broadcast TLB maintenance (BTM) and caches a read-only PTE.

I don't really like this design, but I see how you get here..

mmu notifiers behavior should not be tied to the architecture, they
are supposed to be generic reflections of what the MM is doing so that
they can be hooked into by general purpose drivers.

If you want to hardwire invalidate_range to be only for SVA cases that
actually share the page table itself and rely on some arch-defined
invalidation, then we should give the op a much better name and
discourage anyone else from abusing the new ops variable behavior.

> As no notification is sent and the SMMU does not snoop TLB invalidates
> it will continue to return read-only entries to a device even though
> the CPU page table contains a writable entry. This leads to a
> continually faulting device and no way of handling the fault.

Doesn't the fault generate a PRI/etc? If we get a PRI maybe we should
just have the iommu driver push an iotlb invalidation command before
it acks it? PRI is already really slow so I'm not sure a pipelined
invalidation is going to be a problem? Does the SMMU architecture
permit negative caching which would suggest we need it anyhow?

Jason

WARNING: multiple messages have this Message-ID (diff)
From: Jason Gunthorpe <jgg@nvidia.com>
To: Alistair Popple <apopple@nvidia.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	will@kernel.org, catalin.marinas@arm.com, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org, robin.murphy@arm.com,
	nicolinc@nvidia.com, linux-arm-kernel@lists.infradead.org,
	kvm@vger.kernel.org, John Hubbard <jhubbard@nvidia.com>,
	zhi.wang.linux@gmail.com, Sean Christopherson <seanjc@google.com>
Subject: Re: [PATCH 2/2] arm64: Notify on pte permission upgrades
Date: Sat, 27 May 2023 21:02:13 -0300	[thread overview]
Message-ID: <ZHKaBQt8623s9+VK@nvidia.com> (raw)
In-Reply-To: <5d8e1f752051173d2d1b5c3e14b54eb3506ed3ef.1684892404.git-series.apopple@nvidia.com>

On Wed, May 24, 2023 at 11:47:29AM +1000, Alistair Popple wrote:
> ARM64 requires TLB invalidates when upgrading pte permission from
> read-only to read-write. However mmu_notifiers assume upgrades do not
> need notifications and none are sent. This causes problems when a
> secondary TLB such as implemented by an ARM SMMU doesn't support
> broadcast TLB maintenance (BTM) and caches a read-only PTE.

I don't really like this design, but I see how you get here..

mmu notifiers behavior should not be tied to the architecture, they
are supposed to be generic reflections of what the MM is doing so that
they can be hooked into by general purpose drivers.

If you want to hardwire invalidate_range to be only for SVA cases that
actually share the page table itself and rely on some arch-defined
invalidation, then we should give the op a much better name and
discourage anyone else from abusing the new ops variable behavior.

> As no notification is sent and the SMMU does not snoop TLB invalidates
> it will continue to return read-only entries to a device even though
> the CPU page table contains a writable entry. This leads to a
> continually faulting device and no way of handling the fault.

Doesn't the fault generate a PRI/etc? If we get a PRI maybe we should
just have the iommu driver push an iotlb invalidation command before
it acks it? PRI is already really slow so I'm not sure a pipelined
invalidation is going to be a problem? Does the SMMU architecture
permit negative caching which would suggest we need it anyhow?

Jason

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2023-05-28  0:02 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-24  1:47 [PATCH 1/2] mmu_notifiers: Restore documentation for .invalidate_range() Alistair Popple
2023-05-24  1:47 ` Alistair Popple
2023-05-24  1:47 ` [PATCH 2/2] arm64: Notify on pte permission upgrades Alistair Popple
2023-05-24  1:47   ` Alistair Popple
2023-05-28  0:02   ` Jason Gunthorpe [this message]
2023-05-28  0:02     ` Jason Gunthorpe
2023-05-30  8:05     ` Alistair Popple
2023-05-30  8:05       ` Alistair Popple
2023-05-30 11:54       ` Jason Gunthorpe
2023-05-30 11:54         ` Jason Gunthorpe
2023-05-30 12:14         ` Robin Murphy
2023-05-30 12:14           ` Robin Murphy
2023-05-30 12:52           ` Jason Gunthorpe
2023-05-30 12:52             ` Jason Gunthorpe
2023-05-30 13:44             ` Robin Murphy
2023-05-30 13:44               ` Robin Murphy
2023-05-30 14:06               ` Jason Gunthorpe
2023-05-30 14:06                 ` Jason Gunthorpe
2023-05-30 21:44                 ` Sean Christopherson
2023-05-30 21:44                   ` Sean Christopherson
2023-05-30 23:08                   ` Jason Gunthorpe
2023-05-30 23:08                     ` Jason Gunthorpe
2023-05-31  0:30                     ` Alistair Popple
2023-05-31  0:30                       ` Alistair Popple
2023-05-31  0:32                       ` Jason Gunthorpe
2023-05-31  0:32                         ` Jason Gunthorpe
2023-05-31  2:46                         ` Alistair Popple
2023-05-31  2:46                           ` Alistair Popple
2023-05-31 15:30                           ` Jason Gunthorpe
2023-05-31 15:30                             ` Jason Gunthorpe
2023-05-31 23:56                             ` Alistair Popple
2023-05-31 23:56                               ` Alistair Popple
     [not found]                       ` <31cdd164783fefad4c9ef4a6d33c1e0094405d0f03added523a82dd9febdf15f@mu.id>
2023-06-09  2:06                         ` Alistair Popple
2023-06-09  2:06                           ` Alistair Popple
2023-06-09  6:05                           ` Alistair Popple
2023-06-09  6:05                             ` Alistair Popple
2023-05-24  2:20 ` [PATCH 1/2] mmu_notifiers: Restore documentation for .invalidate_range() John Hubbard
2023-05-24  2:20   ` John Hubbard
2023-05-24  4:45   ` Alistair Popple
2023-05-24  4:45     ` Alistair Popple
2023-05-27 23:56   ` Jason Gunthorpe
2023-05-27 23:56     ` Jason Gunthorpe
2023-05-24  3:48 ` Zhi Wang
2023-05-24  3:48   ` Zhi Wang
2023-05-24  4:57   ` Alistair Popple
2023-05-24  4:57     ` Alistair Popple

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=ZHKaBQt8623s9+VK@nvidia.com \
    --to=jgg@nvidia.com \
    --cc=akpm@linux-foundation.org \
    --cc=apopple@nvidia.com \
    --cc=catalin.marinas@arm.com \
    --cc=jhubbard@nvidia.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=nicolinc@nvidia.com \
    --cc=robin.murphy@arm.com \
    --cc=seanjc@google.com \
    --cc=will@kernel.org \
    --cc=zhi.wang.linux@gmail.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.