Re: [PATCH v2] drm/rockchip: shutdown drm subsystem on shutdown

2019-04-18 Thread Heiko Stuebner
Am Dienstag, 2. April 2019, 13:37:53 CEST schrieb Heiko Stuebner: > From: Vicente Bergas > > As explained by Robin Murphy: > > the IOMMU shutdown disables paging, so if the VOP is still > > scanning out then that will result in whatever IOVAs it was using now going > > straight out onto the bus a

Re: [PATCH v2] drm/rockchip: shutdown drm subsystem on shutdown

2019-04-18 Thread Marc Zyngier
Hi Heiko, On 02/04/2019 12:37, Heiko Stuebner wrote: > From: Vicente Bergas > > As explained by Robin Murphy: >> the IOMMU shutdown disables paging, so if the VOP is still >> scanning out then that will result in whatever IOVAs it was using now going >> straight out onto the bus as physical addr

Re: [PATCH v2] drm/rockchip: shutdown drm subsystem on shutdown

2019-04-12 Thread Doug Anderson
Hi, On Tue, Apr 2, 2019 at 4:38 AM Heiko Stuebner wrote: > > From: Vicente Bergas > > As explained by Robin Murphy: > > the IOMMU shutdown disables paging, so if the VOP is still > > scanning out then that will result in whatever IOVAs it was using now going > > straight out onto the bus as phys

Re: [PATCH v2] drm/rockchip: shutdown drm subsystem on shutdown

2019-04-11 Thread Brian Norris
On Tue, Apr 2, 2019 at 4:38 AM Heiko Stuebner wrote: > > From: Vicente Bergas > > As explained by Robin Murphy: > > the IOMMU shutdown disables paging, so if the VOP is still > > scanning out then that will result in whatever IOVAs it was using now going > > straight out onto the bus as physical

[PATCH v2] drm/rockchip: shutdown drm subsystem on shutdown

2019-04-02 Thread Heiko Stuebner
From: Vicente Bergas As explained by Robin Murphy: > the IOMMU shutdown disables paging, so if the VOP is still > scanning out then that will result in whatever IOVAs it was using now going > straight out onto the bus as physical addresses. We had a more radical approach before in commit 7f3ef5d