All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nicolin Chen <nicolinc@nvidia.com>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: <iommu@lists.linux.dev>, Joerg Roedel <joro@8bytes.org>,
	<linux-arm-kernel@lists.infradead.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Will Deacon <will@kernel.org>,
	Michael Shavit <mshavit@google.com>,
	Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
Subject: Re: [PATCH v2 11/19] iommu/arm-smmu-v3: Do not change the STE twice during arm_smmu_attach_dev()
Date: Mon, 4 Dec 2023 18:46:01 -0800	[thread overview]
Message-ID: <ZW6ObRW5BWbGH7RL@Asurada-Nvidia> (raw)
In-Reply-To: <11-v2-de8b10590bf5+400-smmuv3_newapi_p1_jgg@nvidia.com>

On Mon, Nov 13, 2023 at 01:53:18PM -0400, Jason Gunthorpe wrote:
> This was needed because the STE code required the STE to be in
> ABORT/BYPASS inorder to program a cdtable or S2 STE. Now that the STE code
> can automatically handle all transitions we can remove this step
> from the attach_dev flow.
> 
> A few small bugs exist because of this:
> 
> 1) If the core code does BLOCKED -> UNMANAGED with disable_bypass=false
>    then there will be a moment where the STE points at BYPASS. Since
>    this can be done by VFIO/IOMMUFD it is a small security race.
> 
> 2) If the core code does IDENTITY -> DMA then any IOMMU_RESV_DIRECT
>    regions will temporarily become BLOCKED. We'd like drivers to
>    work in a way that allows IOMMU_RESV_DIRECT to be continuously
>    functional during these transitions.
> 
> Make arm_smmu_release_device() put the STE back to the correct
> ABORT/BYPASS setting. Fix a bug where a IOMMU_RESV_DIRECT was ignored on
> this path.
> 
> Notice this subtly depends on the prior arm_smmu_asid_lock change as the
> STE must be put to non-paging before removing the device for the linked
> list to avoid races with arm_smmu_share_asid().
> 
> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>

WARNING: multiple messages have this Message-ID (diff)
From: Nicolin Chen <nicolinc@nvidia.com>
To: Jason Gunthorpe <jgg@nvidia.com>
Cc: <iommu@lists.linux.dev>, Joerg Roedel <joro@8bytes.org>,
	<linux-arm-kernel@lists.infradead.org>,
	Robin Murphy <robin.murphy@arm.com>,
	Will Deacon <will@kernel.org>,
	Michael Shavit <mshavit@google.com>,
	Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
Subject: Re: [PATCH v2 11/19] iommu/arm-smmu-v3: Do not change the STE twice during arm_smmu_attach_dev()
Date: Mon, 4 Dec 2023 18:46:01 -0800	[thread overview]
Message-ID: <ZW6ObRW5BWbGH7RL@Asurada-Nvidia> (raw)
In-Reply-To: <11-v2-de8b10590bf5+400-smmuv3_newapi_p1_jgg@nvidia.com>

On Mon, Nov 13, 2023 at 01:53:18PM -0400, Jason Gunthorpe wrote:
> This was needed because the STE code required the STE to be in
> ABORT/BYPASS inorder to program a cdtable or S2 STE. Now that the STE code
> can automatically handle all transitions we can remove this step
> from the attach_dev flow.
> 
> A few small bugs exist because of this:
> 
> 1) If the core code does BLOCKED -> UNMANAGED with disable_bypass=false
>    then there will be a moment where the STE points at BYPASS. Since
>    this can be done by VFIO/IOMMUFD it is a small security race.
> 
> 2) If the core code does IDENTITY -> DMA then any IOMMU_RESV_DIRECT
>    regions will temporarily become BLOCKED. We'd like drivers to
>    work in a way that allows IOMMU_RESV_DIRECT to be continuously
>    functional during these transitions.
> 
> Make arm_smmu_release_device() put the STE back to the correct
> ABORT/BYPASS setting. Fix a bug where a IOMMU_RESV_DIRECT was ignored on
> this path.
> 
> Notice this subtly depends on the prior arm_smmu_asid_lock change as the
> STE must be put to non-paging before removing the device for the linked
> list to avoid races with arm_smmu_share_asid().
> 
> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>

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

  parent reply	other threads:[~2023-12-05  2:46 UTC|newest]

Thread overview: 158+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-13 17:53 [PATCH v2 00/19] Update SMMUv3 to the modern iommu API (part 1/3) Jason Gunthorpe
2023-11-13 17:53 ` Jason Gunthorpe
2023-11-13 17:53 ` [PATCH v2 01/19] iommu/arm-smmu-v3: Add a type for the STE Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-14 15:06   ` Moritz Fischer
2023-11-14 15:06     ` Moritz Fischer
2023-11-15 11:52     ` Michael Shavit
2023-11-15 11:52       ` Michael Shavit
2023-11-15 13:35       ` Jason Gunthorpe
2023-11-15 13:35         ` Jason Gunthorpe
2023-11-27 16:03   ` Eric Auger
2023-11-27 16:03     ` Eric Auger
2023-11-27 17:42     ` Jason Gunthorpe
2023-11-27 17:42       ` Jason Gunthorpe
2023-11-27 17:51       ` Eric Auger
2023-11-27 17:51         ` Eric Auger
2023-11-27 18:21         ` Jason Gunthorpe
2023-11-27 18:21           ` Jason Gunthorpe
2023-12-05  0:44   ` Nicolin Chen
2023-12-05  0:44     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 02/19] iommu/arm-smmu-v3: Master cannot be NULL in arm_smmu_write_strtab_ent() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-14 15:17   ` Moritz Fischer
2023-11-14 15:17     ` Moritz Fischer
2023-11-15 11:55     ` Michael Shavit
2023-11-15 11:55       ` Michael Shavit
2023-11-27 15:41   ` Eric Auger
2023-11-27 15:41     ` Eric Auger
2023-12-05  0:45   ` Nicolin Chen
2023-12-05  0:45     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 03/19] iommu/arm-smmu-v3: Remove ARM_SMMU_DOMAIN_NESTED Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-14 15:18   ` Moritz Fischer
2023-11-14 15:18     ` Moritz Fischer
2023-11-27 16:35   ` Eric Auger
2023-11-27 16:35     ` Eric Auger
2023-12-05  0:46   ` Nicolin Chen
2023-12-05  0:46     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 04/19] iommu/arm-smmu-v3: Make STE programming independent of the callers Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-12-05  1:38   ` Nicolin Chen
2023-12-05  1:38     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 05/19] iommu/arm-smmu-v3: Consolidate the STE generation for abort/bypass Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 12:17   ` Michael Shavit
2023-11-15 12:17     ` Michael Shavit
2023-12-05  1:43   ` Nicolin Chen
2023-12-05  1:43     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 06/19] iommu/arm-smmu-v3: Move arm_smmu_rmr_install_bypass_ste() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 13:57   ` Michael Shavit
2023-11-15 13:57     ` Michael Shavit
2023-12-05  1:45   ` Nicolin Chen
2023-12-05  1:45     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 07/19] iommu/arm-smmu-v3: Move the STE generation for S1 and S2 domains into functions Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-14 15:24   ` Moritz Fischer
2023-11-14 15:24     ` Moritz Fischer
2023-11-15 14:01   ` Michael Shavit
2023-11-15 14:01     ` Michael Shavit
2023-12-05  1:55   ` Nicolin Chen
2023-12-05  1:55     ` Nicolin Chen
2023-12-05 14:35     ` Jason Gunthorpe
2023-12-05 14:35       ` Jason Gunthorpe
2023-11-13 17:53 ` [PATCH v2 08/19] iommu/arm-smmu-v3: Build the whole STE in arm_smmu_make_s2_domain_ste() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 14:04   ` Michael Shavit
2023-11-15 14:04     ` Michael Shavit
2023-12-05  1:58   ` Nicolin Chen
2023-12-05  1:58     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 09/19] iommu/arm-smmu-v3: Hold arm_smmu_asid_lock during all of attach_dev Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 14:12   ` Michael Shavit
2023-11-15 14:12     ` Michael Shavit
2023-12-05  2:16   ` Nicolin Chen
2023-12-05  2:16     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 10/19] iommu/arm-smmu-v3: Compute the STE only once for each master Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 14:16   ` Michael Shavit
2023-11-15 14:16     ` Michael Shavit
2023-12-05  2:13   ` Nicolin Chen
2023-12-05  2:13     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 11/19] iommu/arm-smmu-v3: Do not change the STE twice during arm_smmu_attach_dev() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 15:15   ` Michael Shavit
2023-11-15 15:15     ` Michael Shavit
2023-11-16 16:28     ` Jason Gunthorpe
2023-11-16 16:28       ` Jason Gunthorpe
2023-12-05  2:46   ` Nicolin Chen [this message]
2023-12-05  2:46     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 12/19] iommu/arm-smmu-v3: Put writing the context descriptor in the right order Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 15:32   ` Michael Shavit
2023-11-15 15:32     ` Michael Shavit
2023-11-16 16:46     ` Jason Gunthorpe
2023-11-16 16:46       ` Jason Gunthorpe
2023-11-17  4:14       ` Michael Shavit
2023-11-17  4:14         ` Michael Shavit
2023-12-05  3:38   ` Nicolin Chen
2023-12-05  3:38     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 13/19] iommu/arm-smmu-v3: Pass smmu_domain to arm_enable/disable_ats() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-12-05  3:56   ` Nicolin Chen
2023-12-05  3:56     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 14/19] iommu/arm-smmu-v3: Remove arm_smmu_master->domain Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-27 17:14   ` Eric Auger
2023-11-27 17:14     ` Eric Auger
2023-11-30 12:03     ` Jason Gunthorpe
2023-11-30 12:03       ` Jason Gunthorpe
2023-12-05 13:25       ` Eric Auger
2023-12-05 13:25         ` Eric Auger
2023-12-05  4:47   ` Nicolin Chen
2023-12-05  4:47     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 15/19] iommu/arm-smmu-v3: Add a global static IDENTITY domain Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 15:50   ` Michael Shavit
2023-11-15 15:50     ` Michael Shavit
2023-12-05  4:28   ` Nicolin Chen
2023-12-05  4:28     ` Nicolin Chen
2023-12-05 14:37     ` Jason Gunthorpe
2023-12-05 14:37       ` Jason Gunthorpe
2023-12-05 17:25       ` Nicolin Chen
2023-12-05 17:25         ` Nicolin Chen
2023-12-05 17:42         ` Jason Gunthorpe
2023-12-05 17:42           ` Jason Gunthorpe
2023-12-05 18:21           ` Nicolin Chen
2023-12-05 18:21             ` Nicolin Chen
2023-12-05 19:03             ` Jason Gunthorpe
2023-12-05 19:03               ` Jason Gunthorpe
2023-11-13 17:53 ` [PATCH v2 16/19] iommu/arm-smmu-v3: Add a global static BLOCKED domain Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 15:57   ` Michael Shavit
2023-11-15 15:57     ` Michael Shavit
2023-11-16 15:44     ` Jason Gunthorpe
2023-11-16 15:44       ` Jason Gunthorpe
2023-12-05  4:05   ` Nicolin Chen
2023-12-05  4:05     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 17/19] iommu/arm-smmu-v3: Use the identity/blocked domain during release Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-12-05  4:07   ` Nicolin Chen
2023-12-05  4:07     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 18/19] iommu/arm-smmu-v3: Pass arm_smmu_domain and arm_smmu_device to finalize Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-11-15 16:02   ` Michael Shavit
2023-11-15 16:02     ` Michael Shavit
2023-12-05  4:42   ` Nicolin Chen
2023-12-05  4:42     ` Nicolin Chen
2023-11-13 17:53 ` [PATCH v2 19/19] iommu/arm-smmu-v3: Convert to domain_alloc_paging() Jason Gunthorpe
2023-11-13 17:53   ` Jason Gunthorpe
2023-12-05  4:40   ` Nicolin Chen
2023-12-05  4:40     ` Nicolin Chen
2023-11-27 16:10 ` [PATCH v2 00/19] Update SMMUv3 to the modern iommu API (part 1/3) Shameerali Kolothum Thodi
2023-11-27 16:10   ` Shameerali Kolothum Thodi
2023-11-27 17:48   ` Jason Gunthorpe
2023-11-27 17:48     ` Jason Gunthorpe
2023-12-05  3:54 ` Nicolin Chen
2023-12-05  3:54   ` Nicolin Chen

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=ZW6ObRW5BWbGH7RL@Asurada-Nvidia \
    --to=nicolinc@nvidia.com \
    --cc=iommu@lists.linux.dev \
    --cc=jgg@nvidia.com \
    --cc=joro@8bytes.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=mshavit@google.com \
    --cc=robin.murphy@arm.com \
    --cc=shameerali.kolothum.thodi@huawei.com \
    --cc=will@kernel.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.