linux-kernel - Re: [PATCH 1/5] iommu: Add mm_get

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 8 Aug 2023 12:02:23 -0300
From:   Jason Gunthorpe <jgg@...pe.ca>
To:     Tina Zhang <tina.zhang@...el.com>
Cc:     Kevin Tian <kevin.tian@...el.com>,
        Lu Baolu <baolu.lu@...ux.intel.com>,
        Michael Shavit <mshavit@...gle.com>, iommu@...ts.linux.dev,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/5] iommu: Add mm_get_pasid() helper function

On Tue, Aug 08, 2023 at 03:49:40PM +0800, Tina Zhang wrote:
> mm_get_pasid() is for getting mm pasid value.
> 
> The motivation is to replace mm->pasid with an iommu private data
> structure that is introduced in a later patch.

Maybe we should start out by calling it what it actually is:

'mm_get_enqcmd_pasid()'

We can't actually have multiple SVA domains with different PASIDs
until the places wrongly calling this are removed :\

eg, I would expect this series to also come with removing
'pasid_private' from the Intel driver.

The mmu_notifier should be placed in the singular iommu_domain that is
the SVA domain for the mm. Drivers should not attempt to de-duplicate
this, the core code will do it like you are showing in this series.

Jason