[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240806170529.GK676757@ziepe.ca>
Date: Tue, 6 Aug 2024 14:05:29 -0300
From: Jason Gunthorpe <jgg@...pe.ca>
To: Lu Baolu <baolu.lu@...ux.intel.com>
Cc: Joerg Roedel <joro@...tes.org>, Will Deacon <will@...nel.org>,
Robin Murphy <robin.murphy@....com>,
Kevin Tian <kevin.tian@...el.com>, iommu@...ts.linux.dev,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 2/7] iommu/vt-d: Remove identity mappings from
si_domain
On Tue, Aug 06, 2024 at 10:39:36AM +0800, Lu Baolu wrote:
> As the driver has enforced DMA domains for devices managed by an IOMMU
> hardware that doesn't support passthrough translation mode, there is no
> need for static identity mappings in the si_domain. Remove the identity
> mapping code to avoid dead code.
>
> Signed-off-by: Lu Baolu <baolu.lu@...ux.intel.com>
> ---
> drivers/iommu/intel/iommu.c | 122 ++----------------------------------
> 1 file changed, 4 insertions(+), 118 deletions(-)
Reviewed-by: Jason Gunthorpe <jgg@...dia.com>
I always wondered why this was so fine grained, I would have expected
it to work on blocks of 512 x 1G, once you allocate a memory for the
1G table level you may as well fill it with identity IOPTEs. The only
optimization point is how many blocks of 512x1G are needed..
Jason
Powered by blists - more mailing lists