| Message ID | 152153226789.8528.13236911976628814785.sendpatchset@little-apple (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | d2b860cb303fdeffa886c20bfbdbfd88cc52c85f |
| Headers | show |
On Tue, Mar 20, 2018 at 04:51:07PM +0900, Magnus Damm wrote: > From: Magnus Damm <damm+renesas@opensource.se> > > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation > for [R-Car V3M] the DS1 bit field should be bit 0. > > Update the ipmmu-main property to make it match the data sheet. Thanks, I have confirmed this. Apologies for this error. AFIK this IP block is currently unused, I wonder if it, ipmmu_mm and ipmmu_rt should have status = "disabled" added. For now I have applied this change to v4.18 with the following tag: Fixes: ce3b52a1595b ("arm64: dts: renesas: r8a77970: Add IPMMU device nodes") > Signed-off-by: Magnus Damm <damm+renesas@opensource.se> > --- > Based on top of renesas-drivers-2018-03-13-v4.16-rc5 > > arch/arm64/boot/dts/renesas/r8a77970.dtsi | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > --- 0001/arch/arm64/boot/dts/renesas/r8a77970.dtsi > +++ work/arch/arm64/boot/dts/renesas/r8a77970.dtsi 2018-03-20 15:26:18.250607110 +0900 > @@ -160,7 +160,7 @@ > ipmmu_ds1: mmu@e7740000 { > compatible = "renesas,ipmmu-r8a77970"; > reg = <0 0xe7740000 0 0x1000>; > - renesas,ipmmu-main = <&ipmmu_mm 1>; > + renesas,ipmmu-main = <&ipmmu_mm 0>; > power-domains = <&sysc R8A77970_PD_ALWAYS_ON>; > #iommu-cells = <1>; > }; >
On Tue, Mar 20, 2018 at 8:51 AM, Magnus Damm <magnus.damm@gmail.com> wrote: > From: Magnus Damm <damm+renesas@opensource.se> > > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation > for [R-Car V3M] the DS1 bit field should be bit 0. > > Update the ipmmu-main property to make it match the data sheet. > > Signed-off-by: Magnus Damm <damm+renesas@opensource.se> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> Gr{oetje,eeting}s, Geert
On Wed, Mar 21, 2018 at 09:51:33AM +0100, Simon Horman wrote: > On Tue, Mar 20, 2018 at 04:51:07PM +0900, Magnus Damm wrote: > > From: Magnus Damm <damm+renesas@opensource.se> > > > > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation > > for [R-Car V3M] the DS1 bit field should be bit 0. > > > > Update the ipmmu-main property to make it match the data sheet. > > Thanks, I have confirmed this. Apologies for this error. > > AFIK this IP block is currently unused, I wonder if it, > ipmmu_mm and ipmmu_rt should have status = "disabled" added. Sorry, I take that back, I now see the nodes used elsewhere in the same DT file. > > For now I have applied this change to v4.18 with the following tag: > > Fixes: ce3b52a1595b ("arm64: dts: renesas: r8a77970: Add IPMMU device > nodes") > > > Signed-off-by: Magnus Damm <damm+renesas@opensource.se> > > --- > > Based on top of renesas-drivers-2018-03-13-v4.16-rc5 > > > > arch/arm64/boot/dts/renesas/r8a77970.dtsi | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > --- 0001/arch/arm64/boot/dts/renesas/r8a77970.dtsi > > +++ work/arch/arm64/boot/dts/renesas/r8a77970.dtsi 2018-03-20 15:26:18.250607110 +0900 > > @@ -160,7 +160,7 @@ > > ipmmu_ds1: mmu@e7740000 { > > compatible = "renesas,ipmmu-r8a77970"; > > reg = <0 0xe7740000 0 0x1000>; > > - renesas,ipmmu-main = <&ipmmu_mm 1>; > > + renesas,ipmmu-main = <&ipmmu_mm 0>; > > power-domains = <&sysc R8A77970_PD_ALWAYS_ON>; > > #iommu-cells = <1>; > > }; > > >
Hi Simon, On Wed, Mar 21, 2018 at 6:57 PM, Simon Horman <horms@verge.net.au> wrote: > On Wed, Mar 21, 2018 at 09:51:33AM +0100, Simon Horman wrote: >> On Tue, Mar 20, 2018 at 04:51:07PM +0900, Magnus Damm wrote: >> > From: Magnus Damm <damm+renesas@opensource.se> >> > >> > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation >> > for [R-Car V3M] the DS1 bit field should be bit 0. >> > >> > Update the ipmmu-main property to make it match the data sheet. >> >> Thanks, I have confirmed this. Apologies for this error. >> >> AFIK this IP block is currently unused, I wonder if it, >> ipmmu_mm and ipmmu_rt should have status = "disabled" added. > > Sorry, I take that back, I now see the nodes used elsewhere in > the same DT file. Thanks for your help! My opinion is that for R-Car Gen3 we don't really need to keep any IPMMU nodes in disabled state in DT. This since the whitelist code in the IPMMU driver will include software policy handling to during runtime make sure only desired devices get enabled. Cheers, / magnus
On Thu, Mar 22, 2018 at 03:07:49PM +0900, Magnus Damm wrote: > Hi Simon, > > On Wed, Mar 21, 2018 at 6:57 PM, Simon Horman <horms@verge.net.au> wrote: > > On Wed, Mar 21, 2018 at 09:51:33AM +0100, Simon Horman wrote: > >> On Tue, Mar 20, 2018 at 04:51:07PM +0900, Magnus Damm wrote: > >> > From: Magnus Damm <damm+renesas@opensource.se> > >> > > >> > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation > >> > for [R-Car V3M] the DS1 bit field should be bit 0. > >> > > >> > Update the ipmmu-main property to make it match the data sheet. > >> > >> Thanks, I have confirmed this. Apologies for this error. > >> > >> AFIK this IP block is currently unused, I wonder if it, > >> ipmmu_mm and ipmmu_rt should have status = "disabled" added. > > > > Sorry, I take that back, I now see the nodes used elsewhere in > > the same DT file. > > Thanks for your help! My opinion is that for R-Car Gen3 we don't > really need to keep any IPMMU nodes in disabled state in DT. This > since the whitelist code in the IPMMU driver will include software > policy handling to during runtime make sure only desired devices get > enabled. Agreed
Hi Magnus, On Thu, Mar 22, 2018 at 7:07 AM, Magnus Damm <magnus.damm@gmail.com> wrote: > On Wed, Mar 21, 2018 at 6:57 PM, Simon Horman <horms@verge.net.au> wrote: >> On Wed, Mar 21, 2018 at 09:51:33AM +0100, Simon Horman wrote: >>> On Tue, Mar 20, 2018 at 04:51:07PM +0900, Magnus Damm wrote: >>> > From: Magnus Damm <damm+renesas@opensource.se> >>> > >>> > Judging by "R-Car-Gen3-rev0.80" IPMMU IMSSTR register documentation >>> > for [R-Car V3M] the DS1 bit field should be bit 0. >>> > >>> > Update the ipmmu-main property to make it match the data sheet. >>> >>> Thanks, I have confirmed this. Apologies for this error. >>> >>> AFIK this IP block is currently unused, I wonder if it, >>> ipmmu_mm and ipmmu_rt should have status = "disabled" added. >> >> Sorry, I take that back, I now see the nodes used elsewhere in >> the same DT file. > > Thanks for your help! My opinion is that for R-Car Gen3 we don't > really need to keep any IPMMU nodes in disabled state in DT. This > since the whitelist code in the IPMMU driver will include software > policy handling to during runtime make sure only desired devices get > enabled. Good, DT == hardware description taking over again. Gr{oetje,eeting}s, Geert
--- 0001/arch/arm64/boot/dts/renesas/r8a77970.dtsi +++ work/arch/arm64/boot/dts/renesas/r8a77970.dtsi 2018-03-20 15:26:18.250607110 +0900 @@ -160,7 +160,7 @@ ipmmu_ds1: mmu@e7740000 { compatible = "renesas,ipmmu-r8a77970"; reg = <0 0xe7740000 0 0x1000>; - renesas,ipmmu-main = <&ipmmu_mm 1>; + renesas,ipmmu-main = <&ipmmu_mm 0>; power-domains = <&sysc R8A77970_PD_ALWAYS_ON>; #iommu-cells = <1>; };