Add dma properties to support GPCDMA for I2C in Tegra 186 and later
chips
Signed-off-by: Akhil R <[email protected]>
---
arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
3 files changed, 96 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
index 59a10fb184f8..3580fbf99091 100644
--- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
@@ -672,6 +672,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C1>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 21>, <&gpcdma 21>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -685,6 +689,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C3>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 23>, <&gpcdma 23>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -702,6 +710,10 @@
pinctrl-names = "default", "idle";
pinctrl-0 = <&state_dpaux1_i2c>;
pinctrl-1 = <&state_dpaux1_off>;
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 26>, <&gpcdma 26>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -733,6 +745,10 @@
pinctrl-names = "default", "idle";
pinctrl-0 = <&state_dpaux_i2c>;
pinctrl-1 = <&state_dpaux_off>;
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 30>, <&gpcdma 30>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -746,6 +762,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C7>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 27>, <&gpcdma 27>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -759,6 +779,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C9>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 31>, <&gpcdma 31>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -1176,6 +1200,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C2>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 22>, <&gpcdma 22>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -1189,6 +1217,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA186_RESET_I2C8>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 0>, <&gpcdma 0>;
+ dma-names = "rx", "tx";
status = "disabled";
};
diff --git a/arch/arm64/boot/dts/nvidia/tegra194.dtsi b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
index d0ed55e5c860..9176c4b27133 100644
--- a/arch/arm64/boot/dts/nvidia/tegra194.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra194.dtsi
@@ -805,6 +805,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA194_RESET_I2C1>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 21>, <&gpcdma 21>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -830,6 +834,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA194_RESET_I2C3>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 23>, <&gpcdma 23>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -847,6 +855,10 @@
pinctrl-0 = <&state_dpaux1_i2c>;
pinctrl-1 = <&state_dpaux1_off>;
pinctrl-names = "default", "idle";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 26>, <&gpcdma 26>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -864,6 +876,10 @@
pinctrl-0 = <&state_dpaux0_i2c>;
pinctrl-1 = <&state_dpaux0_off>;
pinctrl-names = "default", "idle";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 30>, <&gpcdma 30>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -881,6 +897,10 @@
pinctrl-0 = <&state_dpaux2_i2c>;
pinctrl-1 = <&state_dpaux2_off>;
pinctrl-names = "default", "idle";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 27>, <&gpcdma 27>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -898,6 +918,10 @@
pinctrl-0 = <&state_dpaux3_i2c>;
pinctrl-1 = <&state_dpaux3_off>;
pinctrl-names = "default", "idle";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 31>, <&gpcdma 31>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -1565,6 +1589,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA194_RESET_I2C2>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 22>, <&gpcdma 22>;
+ dma-names = "rx", "tx";
status = "disabled";
};
@@ -1578,6 +1606,10 @@
clock-names = "div-clk";
resets = <&bpmp TEGRA194_RESET_I2C8>;
reset-names = "i2c";
+ iommus = <&smmu TEGRA194_SID_GPCDMA_0>;
+ dma-coherent;
+ dmas = <&gpcdma 0>, <&gpcdma 0>;
+ dma-names = "rx", "tx";
status = "disabled";
};
diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
index 81a0f599685f..5852e765ad90 100644
--- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
@@ -737,6 +737,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C1>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 21>, <&gpcdma 21>;
+ dma-names = "rx", "tx";
};
cam_i2c: i2c@3180000 {
@@ -752,6 +756,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C3>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 23>, <&gpcdma 23>;
+ dma-names = "rx", "tx";
};
dp_aux_ch1_i2c: i2c@3190000 {
@@ -767,6 +775,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C4>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 26>, <&gpcdma 26>;
+ dma-names = "rx", "tx";
};
dp_aux_ch0_i2c: i2c@31b0000 {
@@ -782,6 +794,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C6>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 30>, <&gpcdma 30>;
+ dma-names = "rx", "tx";
};
dp_aux_ch2_i2c: i2c@31c0000 {
@@ -797,6 +813,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C7>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 27>, <&gpcdma 27>;
+ dma-names = "rx", "tx";
};
dp_aux_ch3_i2c: i2c@31e0000 {
@@ -812,6 +832,10 @@
clock-names = "div-clk", "parent";
resets = <&bpmp TEGRA234_RESET_I2C9>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 31>, <&gpcdma 31>;
+ dma-names = "rx", "tx";
};
spi@3270000 {
@@ -1109,6 +1133,10 @@
assigned-clock-parents = <&bpmp TEGRA234_CLK_PLLP_OUT0>;
resets = <&bpmp TEGRA234_RESET_I2C2>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 22>, <&gpcdma 22>;
+ dma-names = "rx", "tx";
};
gen8_i2c: i2c@c250000 {
@@ -1125,6 +1153,10 @@
assigned-clock-parents = <&bpmp TEGRA234_CLK_PLLP_OUT0>;
resets = <&bpmp TEGRA234_RESET_I2C8>;
reset-names = "i2c";
+ iommus = <&smmu_niso0 TEGRA234_SID_GPCDMA>;
+ dma-coherent;
+ dmas = <&gpcdma 0>, <&gpcdma 0>;
+ dma-names = "rx", "tx";
};
rtc@c2a0000 {
--
2.17.1
On Tue, Sep 06, 2022 at 08:17:15PM +0530, Akhil R wrote:
> Add dma properties to support GPCDMA for I2C in Tegra 186 and later
> chips
>
> Signed-off-by: Akhil R <[email protected]>
> ---
> arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
> arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
> arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
> 3 files changed, 96 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> index 59a10fb184f8..3580fbf99091 100644
> --- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> +++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> @@ -672,6 +672,10 @@
> clock-names = "div-clk";
> resets = <&bpmp TEGRA186_RESET_I2C1>;
> reset-names = "i2c";
> + iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
> + dma-coherent;
I wonder: why do we need the iommus and dma-coherent properties here?
The I2C controllers are not directly accessing memory, instead it's the
GPCDMA via the dmas/dma-names properties. The GPCDMA already has these
properties set, so they seem to be useless here.
I'm asking because they cause a lot of warnings from the DT validators,
so we either need to remove them (if they are not necessary) or add the
DT binding documentation for them.
Thierry
On Fri, Oct 07, 2022 at 03:20:37PM +0200, Thierry Reding wrote:
> On Tue, Sep 06, 2022 at 08:17:15PM +0530, Akhil R wrote:
> > Add dma properties to support GPCDMA for I2C in Tegra 186 and later
> > chips
> >
> > Signed-off-by: Akhil R <[email protected]>
> > ---
> > arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
> > arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
> > arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
> > 3 files changed, 96 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > index 59a10fb184f8..3580fbf99091 100644
> > --- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > +++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > @@ -672,6 +672,10 @@
> > clock-names = "div-clk";
> > resets = <&bpmp TEGRA186_RESET_I2C1>;
> > reset-names = "i2c";
> > + iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
> > + dma-coherent;
>
> I wonder: why do we need the iommus and dma-coherent properties here?
> The I2C controllers are not directly accessing memory, instead it's the
> GPCDMA via the dmas/dma-names properties. The GPCDMA already has these
> properties set, so they seem to be useless here.
Looking at this some more, the reason why we need these is so that the
struct device backing these I2C controllers is attached to an IOMMU and
the DMA ops are set up correspondingly. Without these, the DMA memory
allocated by the I2C controllers will not be mapped through the IOMMU
and cause faults because the GPCDMA is the one that needs to access
those.
I do recall that we have a similar case for audio where the "sound card"
needs to have an iommus property to make sure it allocates memory
through the same IOMMU domain as the ADMA, which is the device that ends
up doing the actual memory accesses.
Rob, Robin, Will, do you know of a good way other than the DT workaround
here to address this? I think ideally we would want to obtain the "DMA
parent" of these devices so that we allocate memory for that parent
instead of the child. We do have some existing infrastructure for this
type of relationship with the __of_get_dma_parent() function as well as
the interconnects property, but I wonder if that's really the right way
to represent this.
Adding "interconnects" properties would also duplicate the "dmas"
properties we already use to obtain the TX and RX DMA channels. One
simple way to more accurately do this would be to reach into the DMA
engine internals (dma_chan->device->dev) and pass that to dma_alloc_*()
to make sure we allocate for the correct device. For audio that could be
a bit complicated because most of that code is shared across multiple
vendors. I couldn't find any examples where a driver would reach into
DMA channels to allocate for the parent, so I'm wondering what other
people do to solve this issue. Or if anyone else even has the same
issue.
Adding Lars-Peter for the sound/dmaengine helpers and Vinod for general
dmaengine. Perhaps they have some thoughts on or experience with this.
Thierry
On 2022-10-07 15:34, Thierry Reding wrote:
> On Fri, Oct 07, 2022 at 03:20:37PM +0200, Thierry Reding wrote:
>> On Tue, Sep 06, 2022 at 08:17:15PM +0530, Akhil R wrote:
>>> Add dma properties to support GPCDMA for I2C in Tegra 186 and later
>>> chips
>>>
>>> Signed-off-by: Akhil R <[email protected]>
>>> ---
>>> arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
>>> arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
>>> arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
>>> 3 files changed, 96 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>> index 59a10fb184f8..3580fbf99091 100644
>>> --- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>> +++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>> @@ -672,6 +672,10 @@
>>> clock-names = "div-clk";
>>> resets = <&bpmp TEGRA186_RESET_I2C1>;
>>> reset-names = "i2c";
>>> + iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
>>> + dma-coherent;
>>
>> I wonder: why do we need the iommus and dma-coherent properties here?
>> The I2C controllers are not directly accessing memory, instead it's the
>> GPCDMA via the dmas/dma-names properties. The GPCDMA already has these
>> properties set, so they seem to be useless here.
>
> Looking at this some more, the reason why we need these is so that the
> struct device backing these I2C controllers is attached to an IOMMU and
> the DMA ops are set up correspondingly. Without these, the DMA memory
> allocated by the I2C controllers will not be mapped through the IOMMU
> and cause faults because the GPCDMA is the one that needs to access
> those.
>
> I do recall that we have a similar case for audio where the "sound card"
> needs to have an iommus property to make sure it allocates memory
> through the same IOMMU domain as the ADMA, which is the device that ends
> up doing the actual memory accesses.
>
> Rob, Robin, Will, do you know of a good way other than the DT workaround
> here to address this? I think ideally we would want to obtain the "DMA
> parent" of these devices so that we allocate memory for that parent
> instead of the child. We do have some existing infrastructure for this
> type of relationship with the __of_get_dma_parent() function as well as
> the interconnects property, but I wonder if that's really the right way
> to represent this.
>
> Adding "interconnects" properties would also duplicate the "dmas"
> properties we already use to obtain the TX and RX DMA channels. One
> simple way to more accurately do this would be to reach into the DMA
> engine internals (dma_chan->device->dev) and pass that to dma_alloc_*()
> to make sure we allocate for the correct device. For audio that could be
> a bit complicated because most of that code is shared across multiple
> vendors. I couldn't find any examples where a driver would reach into
> DMA channels to allocate for the parent, so I'm wondering what other
> people do to solve this issue. Or if anyone else even has the same
> issue.
As far as I'm aware that's the correct approach, i.e. if a driver is
using an external dmaengine then it's responsible for making DMA
mappings for the correct DMA channel device. We ended up being a bit
asymmetrical in that the dmaengine driver itself has to take care of its
own mapping for the non-memory end of a transfer when an IOMMU is
involved - that's what dma_map_resource() was created for, see pl330 and
rcar-dmac for examples.
The only driver I have first-hand experience with in this context is
amba-pl011, using pl330 through an SMMU on the Arm Juno board, but that
definitely works fine without DT hacks.
Robin.
> Adding Lars-Peter for the sound/dmaengine helpers and Vinod for general
> dmaengine. Perhaps they have some thoughts on or experience with this.
>
> Thierry
On Fri, Oct 07, 2022 at 03:53:21PM +0100, Robin Murphy wrote:
> On 2022-10-07 15:34, Thierry Reding wrote:
> > On Fri, Oct 07, 2022 at 03:20:37PM +0200, Thierry Reding wrote:
> > > On Tue, Sep 06, 2022 at 08:17:15PM +0530, Akhil R wrote:
> > > > Add dma properties to support GPCDMA for I2C in Tegra 186 and later
> > > > chips
> > > >
> > > > Signed-off-by: Akhil R <[email protected]>
> > > > ---
> > > > arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
> > > > arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
> > > > arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
> > > > 3 files changed, 96 insertions(+)
> > > >
> > > > diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > > > index 59a10fb184f8..3580fbf99091 100644
> > > > --- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > > > +++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
> > > > @@ -672,6 +672,10 @@
> > > > clock-names = "div-clk";
> > > > resets = <&bpmp TEGRA186_RESET_I2C1>;
> > > > reset-names = "i2c";
> > > > + iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
> > > > + dma-coherent;
> > >
> > > I wonder: why do we need the iommus and dma-coherent properties here?
> > > The I2C controllers are not directly accessing memory, instead it's the
> > > GPCDMA via the dmas/dma-names properties. The GPCDMA already has these
> > > properties set, so they seem to be useless here.
> >
> > Looking at this some more, the reason why we need these is so that the
> > struct device backing these I2C controllers is attached to an IOMMU and
> > the DMA ops are set up correspondingly. Without these, the DMA memory
> > allocated by the I2C controllers will not be mapped through the IOMMU
> > and cause faults because the GPCDMA is the one that needs to access
> > those.
> >
> > I do recall that we have a similar case for audio where the "sound card"
> > needs to have an iommus property to make sure it allocates memory
> > through the same IOMMU domain as the ADMA, which is the device that ends
> > up doing the actual memory accesses.
> >
> > Rob, Robin, Will, do you know of a good way other than the DT workaround
> > here to address this? I think ideally we would want to obtain the "DMA
> > parent" of these devices so that we allocate memory for that parent
> > instead of the child. We do have some existing infrastructure for this
> > type of relationship with the __of_get_dma_parent() function as well as
> > the interconnects property, but I wonder if that's really the right way
> > to represent this.
> >
> > Adding "interconnects" properties would also duplicate the "dmas"
> > properties we already use to obtain the TX and RX DMA channels. One
> > simple way to more accurately do this would be to reach into the DMA
> > engine internals (dma_chan->device->dev) and pass that to dma_alloc_*()
> > to make sure we allocate for the correct device. For audio that could be
> > a bit complicated because most of that code is shared across multiple
> > vendors. I couldn't find any examples where a driver would reach into
> > DMA channels to allocate for the parent, so I'm wondering what other
> > people do to solve this issue. Or if anyone else even has the same
> > issue.
>
> As far as I'm aware that's the correct approach, i.e. if a driver is using
> an external dmaengine then it's responsible for making DMA mappings for the
> correct DMA channel device. We ended up being a bit asymmetrical in that the
> dmaengine driver itself has to take care of its own mapping for the
> non-memory end of a transfer when an IOMMU is involved - that's what
> dma_map_resource() was created for, see pl330 and rcar-dmac for examples.
>
> The only driver I have first-hand experience with in this context is
> amba-pl011, using pl330 through an SMMU on the Arm Juno board, but that
> definitely works fine without DT hacks.
Oh yeah, I see. That's exactly what I had in mind. And now that I'm
using the right regex I can also find more occurrences of this. Thanks
for those pointers. Looks like that's the right approach then. I'll try
to make corresponding changes and see if there's any fallout.
Thanks,
Thierry
>
> Robin.
>
> > Adding Lars-Peter for the sound/dmaengine helpers and Vinod for general
> > dmaengine. Perhaps they have some thoughts on or experience with this.
> >
> > Thierry
On 2022-10-07 16:17, Thierry Reding wrote:
> On Fri, Oct 07, 2022 at 03:53:21PM +0100, Robin Murphy wrote:
>> On 2022-10-07 15:34, Thierry Reding wrote:
>>> On Fri, Oct 07, 2022 at 03:20:37PM +0200, Thierry Reding wrote:
>>>> On Tue, Sep 06, 2022 at 08:17:15PM +0530, Akhil R wrote:
>>>>> Add dma properties to support GPCDMA for I2C in Tegra 186 and later
>>>>> chips
>>>>>
>>>>> Signed-off-by: Akhil R <[email protected]>
>>>>> ---
>>>>> arch/arm64/boot/dts/nvidia/tegra186.dtsi | 32 ++++++++++++++++++++++++
>>>>> arch/arm64/boot/dts/nvidia/tegra194.dtsi | 32 ++++++++++++++++++++++++
>>>>> arch/arm64/boot/dts/nvidia/tegra234.dtsi | 32 ++++++++++++++++++++++++
>>>>> 3 files changed, 96 insertions(+)
>>>>>
>>>>> diff --git a/arch/arm64/boot/dts/nvidia/tegra186.dtsi b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>>>> index 59a10fb184f8..3580fbf99091 100644
>>>>> --- a/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>>>> +++ b/arch/arm64/boot/dts/nvidia/tegra186.dtsi
>>>>> @@ -672,6 +672,10 @@
>>>>> clock-names = "div-clk";
>>>>> resets = <&bpmp TEGRA186_RESET_I2C1>;
>>>>> reset-names = "i2c";
>>>>> + iommus = <&smmu TEGRA186_SID_GPCDMA_0>;
>>>>> + dma-coherent;
>>>>
>>>> I wonder: why do we need the iommus and dma-coherent properties here?
>>>> The I2C controllers are not directly accessing memory, instead it's the
>>>> GPCDMA via the dmas/dma-names properties. The GPCDMA already has these
>>>> properties set, so they seem to be useless here.
>>>
>>> Looking at this some more, the reason why we need these is so that the
>>> struct device backing these I2C controllers is attached to an IOMMU and
>>> the DMA ops are set up correspondingly. Without these, the DMA memory
>>> allocated by the I2C controllers will not be mapped through the IOMMU
>>> and cause faults because the GPCDMA is the one that needs to access
>>> those.
>>>
>>> I do recall that we have a similar case for audio where the "sound card"
>>> needs to have an iommus property to make sure it allocates memory
>>> through the same IOMMU domain as the ADMA, which is the device that ends
>>> up doing the actual memory accesses.
>>>
>>> Rob, Robin, Will, do you know of a good way other than the DT workaround
>>> here to address this? I think ideally we would want to obtain the "DMA
>>> parent" of these devices so that we allocate memory for that parent
>>> instead of the child. We do have some existing infrastructure for this
>>> type of relationship with the __of_get_dma_parent() function as well as
>>> the interconnects property, but I wonder if that's really the right way
>>> to represent this.
>>>
>>> Adding "interconnects" properties would also duplicate the "dmas"
>>> properties we already use to obtain the TX and RX DMA channels. One
>>> simple way to more accurately do this would be to reach into the DMA
>>> engine internals (dma_chan->device->dev) and pass that to dma_alloc_*()
>>> to make sure we allocate for the correct device. For audio that could be
>>> a bit complicated because most of that code is shared across multiple
>>> vendors. I couldn't find any examples where a driver would reach into
>>> DMA channels to allocate for the parent, so I'm wondering what other
>>> people do to solve this issue. Or if anyone else even has the same
>>> issue.
>>
>> As far as I'm aware that's the correct approach, i.e. if a driver is using
>> an external dmaengine then it's responsible for making DMA mappings for the
>> correct DMA channel device. We ended up being a bit asymmetrical in that the
>> dmaengine driver itself has to take care of its own mapping for the
>> non-memory end of a transfer when an IOMMU is involved - that's what
>> dma_map_resource() was created for, see pl330 and rcar-dmac for examples.
>>
>> The only driver I have first-hand experience with in this context is
>> amba-pl011, using pl330 through an SMMU on the Arm Juno board, but that
>> definitely works fine without DT hacks.
>
> Oh yeah, I see. That's exactly what I had in mind. And now that I'm
> using the right regex I can also find more occurrences of this. Thanks
> for those pointers. Looks like that's the right approach then. I'll try
> to make corresponding changes and see if there's any fallout.
Eww, from a quick look the DMA API usage in i2c-tegra is worse than just
this. Calling dma_sync_* on a buffer from dma_alloc_coherent() is both
egregiously wrong and semantically pointless - the fundamental purpose
of a coherent allocation is that both CPU and device can access it at
any time without any explicit synchronisation. If the driver's doing its
own bounce-buffering for reasons of data alignment then the syncs should
just go; otherwise get rid of the coherent buffer and replace the syncs
with proper map/unmap. Either way, be sure to throw CONFIG_DMA_API_DEBUG
at it, hard.
Cheers,
Robin.