[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250902130750.GG186519@nvidia.com>
Date: Tue, 2 Sep 2025 10:07:50 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: Suravee Suthikulpanit <suravee.suthikulpanit@....com>
Cc: nicolinc@...dia.com, linux-kernel@...r.kernel.org, robin.murphy@....com,
will@...nel.org, joro@...tes.org, kevin.tian@...el.com,
jsnitsel@...hat.com, vasant.hegde@....com, iommu@...ts.linux.dev,
santosh.shukla@....com, sairaj.arunkodilkar@....com,
jon.grimm@....com, prashanthpra@...gle.com, wvw@...gle.com,
wnliu@...gle.com, gptran@...gle.com, kpsingh@...gle.com
Subject: Re: [PATCH 4/8] iommu/amd: Introduce struct gcr3_tbl_info.giov
On Wed, Aug 20, 2025 at 11:30:05AM +0000, Suravee Suthikulpanit wrote:
> @@ -2036,7 +2037,7 @@ static void set_dte_gcr3_table(struct amd_iommu *iommu,
> target->data[0] |= DTE_FLAG_GV |
> FIELD_PREP(DTE_GLX, gcr3_info->glx) |
> FIELD_PREP(DTE_GCR3_14_12, gcr3 >> 12);
> - if (pdom_is_v2_pgtbl_mode(dev_data->domain))
> + if (gcr3_info->giov)
> target->data[0] |= DTE_FLAG_GIOV;
>
> target->data[1] |= FIELD_PREP(DTE_GCR3_30_15, gcr3 >> 15) |
Can we please fix this properly as I've asked many times now and given
sample code :\
Building the DTE for nested should be done with a more natural
connection to the provided vDTE fragment, not by trying to squeeze it
through this existing function by re-using gcr3_info.
Jason
Powered by blists - more mailing lists