All of lore.kernel.org
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: iommu@lists.linux-foundation.org
Cc: linux-kernel@vger.kernel.org, Will Deacon <will@kernel.org>,
	Guenter Roeck <linux@roeck-us.net>,
	Claire Chang <tientzu@chromium.org>,
	Christoph Hellwig <hch@lst.de>,
	Robin Murphy <robin.murphy@arm.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Nathan Chancellor <nathan@kernel.org>
Subject: [PATCH 0/5] Fix restricted DMA vs swiotlb_exit()
Date: Mon, 19 Jul 2021 13:30:49 +0100	[thread overview]
Message-ID: <20210719123054.6844-1-will@kernel.org> (raw)

Hi all,

This series fixes the issues which have been reported against the
Restricted DMA series in -next:

  * Fix the build for Sparc as reported by Guenter [1].

  * Rework the lifetime of 'io_tlb_default_mem' so that devices
    can retain valid references to it even after swiotlb_exit(). This,
    in turn, fixes the x86/AMD IOMMU regressions reported by Nathan [2].

I also then added a diagnostic to swiotlb_exit(), as suggested by Konrad
[3] and the final patch frees the underlying buffer memory during the
tear down, but I must confess that I don't know why this wasn't being
done already.

A massive thank you to Nathan for helping to debug this and also for
testing these patches to confirm that they address the issue on his
machine.

Patches are based against swiotlb devel/for-linus-5.15.

Cheers,

Will

[1] https://lore.kernel.org/r/20210702030807.GA2685166@roeck-us.net
[2] https://lore.kernel.org/r/YNvMDFWKXSm4LRfZ@Ryzen-9-3900X.localdomain
[3] https://lore.kernel.org/r/YORsr0h7u5l9DZwh@char.us.oracle.com

Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Claire Chang <tientzu@chromium.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Nathan Chancellor <nathan@kernel.org>

--->8

Will Deacon (5):
  of: Return success from of_dma_set_restricted_buffer() when
    !OF_ADDRESS
  swiotlb: Point io_default_tlb_mem at static allocation
  swiotlb: Remove io_tlb_default_mem indirection
  swiotlb: Emit diagnostic in swiotlb_exit()
  swiotlb: Free tbl memory in swiotlb_exit()

 drivers/base/core.c       |  2 +-
 drivers/of/of_private.h   |  3 +-
 drivers/xen/swiotlb-xen.c |  4 +-
 include/linux/swiotlb.h   |  4 +-
 kernel/dma/swiotlb.c      | 82 +++++++++++++++++++++++----------------
 5 files changed, 56 insertions(+), 39 deletions(-)

-- 
2.32.0.402.g57bb445576-goog


WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will@kernel.org>
To: iommu@lists.linux-foundation.org
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Will Deacon <will@kernel.org>,
	linux-kernel@vger.kernel.org,
	Nathan Chancellor <nathan@kernel.org>,
	Claire Chang <tientzu@chromium.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Christoph Hellwig <hch@lst.de>,
	Guenter Roeck <linux@roeck-us.net>
Subject: [PATCH 0/5] Fix restricted DMA vs swiotlb_exit()
Date: Mon, 19 Jul 2021 13:30:49 +0100	[thread overview]
Message-ID: <20210719123054.6844-1-will@kernel.org> (raw)

Hi all,

This series fixes the issues which have been reported against the
Restricted DMA series in -next:

  * Fix the build for Sparc as reported by Guenter [1].

  * Rework the lifetime of 'io_tlb_default_mem' so that devices
    can retain valid references to it even after swiotlb_exit(). This,
    in turn, fixes the x86/AMD IOMMU regressions reported by Nathan [2].

I also then added a diagnostic to swiotlb_exit(), as suggested by Konrad
[3] and the final patch frees the underlying buffer memory during the
tear down, but I must confess that I don't know why this wasn't being
done already.

A massive thank you to Nathan for helping to debug this and also for
testing these patches to confirm that they address the issue on his
machine.

Patches are based against swiotlb devel/for-linus-5.15.

Cheers,

Will

[1] https://lore.kernel.org/r/20210702030807.GA2685166@roeck-us.net
[2] https://lore.kernel.org/r/YNvMDFWKXSm4LRfZ@Ryzen-9-3900X.localdomain
[3] https://lore.kernel.org/r/YORsr0h7u5l9DZwh@char.us.oracle.com

Cc: Guenter Roeck <linux@roeck-us.net>
Cc: Claire Chang <tientzu@chromium.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: Nathan Chancellor <nathan@kernel.org>

--->8

Will Deacon (5):
  of: Return success from of_dma_set_restricted_buffer() when
    !OF_ADDRESS
  swiotlb: Point io_default_tlb_mem at static allocation
  swiotlb: Remove io_tlb_default_mem indirection
  swiotlb: Emit diagnostic in swiotlb_exit()
  swiotlb: Free tbl memory in swiotlb_exit()

 drivers/base/core.c       |  2 +-
 drivers/of/of_private.h   |  3 +-
 drivers/xen/swiotlb-xen.c |  4 +-
 include/linux/swiotlb.h   |  4 +-
 kernel/dma/swiotlb.c      | 82 +++++++++++++++++++++++----------------
 5 files changed, 56 insertions(+), 39 deletions(-)

-- 
2.32.0.402.g57bb445576-goog

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

             reply	other threads:[~2021-07-19 12:31 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-19 12:30 Will Deacon [this message]
2021-07-19 12:30 ` [PATCH 0/5] Fix restricted DMA vs swiotlb_exit() Will Deacon
2021-07-19 12:30 ` [PATCH 1/5] of: Return success from of_dma_set_restricted_buffer() when !OF_ADDRESS Will Deacon
2021-07-19 12:30   ` Will Deacon
2021-07-20  3:35   ` Claire Chang
2021-07-20  3:35     ` Claire Chang
2021-07-19 12:30 ` [PATCH 2/5] swiotlb: Point io_default_tlb_mem at static allocation Will Deacon
2021-07-19 12:30   ` Will Deacon
2021-07-20  3:35   ` Claire Chang
2021-07-20  3:35     ` Claire Chang
2021-07-20  7:51   ` Christoph Hellwig
2021-07-20  7:51     ` Christoph Hellwig
2021-07-20  8:49     ` Will Deacon
2021-07-20  8:49       ` Will Deacon
2021-07-19 12:30 ` [PATCH 3/5] swiotlb: Remove io_tlb_default_mem indirection Will Deacon
2021-07-19 12:30   ` Will Deacon
2021-07-20  3:35   ` Claire Chang
2021-07-20  3:35     ` Claire Chang
2021-07-19 12:30 ` [PATCH 4/5] swiotlb: Emit diagnostic in swiotlb_exit() Will Deacon
2021-07-19 12:30   ` Will Deacon
2021-07-20  3:36   ` Claire Chang
2021-07-20  3:36     ` Claire Chang
2021-07-19 12:30 ` [PATCH 5/5] swiotlb: Free tbl memory " Will Deacon
2021-07-19 12:30   ` Will Deacon
2021-07-20  3:36   ` Claire Chang
2021-07-20  3:36     ` Claire Chang
2021-07-20  8:35     ` Will Deacon
2021-07-20  8:35       ` Will Deacon
2021-07-20  7:51 ` [PATCH 0/5] Fix restricted DMA vs swiotlb_exit() Christoph Hellwig
2021-07-20  7:51   ` Christoph Hellwig

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=20210719123054.6844-1-will@kernel.org \
    --to=will@kernel.org \
    --cc=hch@lst.de \
    --cc=iommu@lists.linux-foundation.org \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@roeck-us.net \
    --cc=nathan@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=tientzu@chromium.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 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.