[REGRESSION] on linux-next (next-20250507)
Borah, Chaitanya Kumar
chaitanya.kumar.borah at intel.com
Wed May 14 12:40:02 UTC 2025
> -----Original Message-----
> From: Jason Gunthorpe <jgg at nvidia.com>
> Sent: Monday, May 12, 2025 5:51 PM
> To: Borah, Chaitanya Kumar <chaitanya.kumar.borah at intel.com>
> Cc: intel-xe at lists.freedesktop.org; intel-gfx at lists.freedesktop.org; Saarinen,
> Jani <jani.saarinen at intel.com>; Nikula, Jani <jani.nikula at intel.com>; Kurmi,
> Suresh Kumar <suresh.kumar.kurmi at intel.com>; De Marchi, Lucas
> <lucas.demarchi at intel.com>; iommu at lists.linux.dev
> Subject: Re: [REGRESSION] on linux-next (next-20250507)
>
> On Mon, May 12, 2025 at 06:00:39AM +0000, Borah, Chaitanya Kumar
> wrote:
>
> > ``````````````````````````````````````````````````````````````````````
> > ```````````````````````````````````
> > commit 13f43d7cf3e0570004a0d960bc1be23db827c2ff
> > Author: Jason Gunthorpe mailto:jgg at nvidia.com
> > Date: Tue Apr 8 13:53:56 2025 -0300
> >
> > iommu/pages: Formalize the freelist API
> > ``````````````````````````````````````````````````````````````````````
> > ```````````````````````````````````
> >
> > We could not revert the patch because of merge conflicts but resetting to the
> parent of the commit seems to fix the issue.
> >
> > Could you please check why the patch causes this regression and provide a
> fix if necessary?
>
> Could this be the solution?
Apologies for the delay, this change seems to solve the issue. Can we expect a patch soon?
Regards
Chaitanya
>
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index 0af1ab36283cba..7d2b51a890c75a 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -154,6 +154,8 @@ static void fq_ring_free_locked(struct
> iommu_dma_cookie *cookie, struct iova_fq
> fq->entries[idx].iova_pfn,
> fq->entries[idx].pages);
>
> + fq->entries[idx].freelist =
> + IOMMU_PAGES_LIST_INIT(fq->entries[idx].freelist);
> fq->head = (fq->head + 1) & fq->mod_mask;
> }
> }
>
> Jason
More information about the Intel-gfx
mailing list