diff mbox series

[1/1] arm64: dts: imx8mp: remove arm,primecell-periphid at etm nodes

Message ID 20230705205954.4159781-1-Frank.Li@nxp.com
State Accepted
Commit ba345b77fae7054d0cbd033283c47033e45db6d8
Headers show
Series [1/1] arm64: dts: imx8mp: remove arm,primecell-periphid at etm nodes | expand

Commit Message

Frank Li July 5, 2023, 8:59 p.m. UTC
The reg size of etm nodes is incorrectly set to 64k instead of 4k. This
leads to a crash when calling amba_read_periphid().  After corrected reg
size, amba_read_periphid() retrieve the correct periphid.
arm,primecell-periphid were removed from the etm nodes.

Signed-off-by: Frank Li <Frank.Li@nxp.com>
---
 arch/arm64/boot/dts/freescale/imx8mp.dtsi | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

Comments

Alexander Stein July 6, 2023, 5:06 a.m. UTC | #1
Hi Frank,

Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> The reg size of etm nodes is incorrectly set to 64k instead of 4k. This
> leads to a crash when calling amba_read_periphid().  After corrected reg
> size, amba_read_periphid() retrieve the correct periphid.
> arm,primecell-periphid were removed from the etm nodes.

So this means the reference manual is wrong here? It clearly states the size 
is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
On a side note: Is imx8mq affected by this as well? The DAP memory table lists 
similar sizes in the RM .

Best regards,
Alexander

> Signed-off-by: Frank Li <Frank.Li@nxp.com>
> ---
>  arch/arm64/boot/dts/freescale/imx8mp.dtsi | 12 ++++--------
>  1 file changed, 4 insertions(+), 8 deletions(-)
> 
> diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> b/arch/arm64/boot/dts/freescale/imx8mp.dtsi index
> cc406bb338fe..e0ca82ff6f15 100644
> --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> @@ -306,8 +306,7 @@ soc: soc@0 {
> 
>  		etm0: etm@28440000 {
>  			compatible = "arm,coresight-etm4x", 
"arm,primecell";
> -			reg = <0x28440000 0x10000>;
> -			arm,primecell-periphid = <0xbb95d>;
> +			reg = <0x28440000 0x1000>;
>  			cpu = <&A53_0>;
>  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>  			clock-names = "apb_pclk";
> @@ -323,8 +322,7 @@ etm0_out_port: endpoint {
> 
>  		etm1: etm@28540000 {
>  			compatible = "arm,coresight-etm4x", 
"arm,primecell";
> -			reg = <0x28540000 0x10000>;
> -			arm,primecell-periphid = <0xbb95d>;
> +			reg = <0x28540000 0x1000>;
>  			cpu = <&A53_1>;
>  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>  			clock-names = "apb_pclk";
> @@ -340,8 +338,7 @@ etm1_out_port: endpoint {
> 
>  		etm2: etm@28640000 {
>  			compatible = "arm,coresight-etm4x", 
"arm,primecell";
> -			reg = <0x28640000 0x10000>;
> -			arm,primecell-periphid = <0xbb95d>;
> +			reg = <0x28640000 0x1000>;
>  			cpu = <&A53_2>;
>  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>  			clock-names = "apb_pclk";
> @@ -357,8 +354,7 @@ etm2_out_port: endpoint {
> 
>  		etm3: etm@28740000 {
>  			compatible = "arm,coresight-etm4x", 
"arm,primecell";
> -			reg = <0x28740000 0x10000>;
> -			arm,primecell-periphid = <0xbb95d>;
> +			reg = <0x28740000 0x1000>;
>  			cpu = <&A53_3>;
>  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>  			clock-names = "apb_pclk";
Lucas Stach July 6, 2023, 8:23 a.m. UTC | #2
Hi Alexander,

Am Donnerstag, dem 06.07.2023 um 07:06 +0200 schrieb Alexander Stein:
> Hi Frank,
> 
> Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> > The reg size of etm nodes is incorrectly set to 64k instead of 4k. This
> > leads to a crash when calling amba_read_periphid().  After corrected reg
> > size, amba_read_periphid() retrieve the correct periphid.
> > arm,primecell-periphid were removed from the etm nodes.
> 
> So this means the reference manual is wrong here? It clearly states the size 
> is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
> On a side note: Is imx8mq affected by this as well? The DAP memory table lists 
> similar sizes in the RM .
> 
Note that the 64K MMIO space per device is really an alignment thing.
It's a recommendation from ARM to allow individual device MMIO regions
to be mapped on kernels with 64K page size. Most of the time the real
MMIO space occupied by the device is actually much smaller than 64K.

Regards,
Lucas

> Best regards,
> Alexander
> 
> > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > ---
> >  arch/arm64/boot/dts/freescale/imx8mp.dtsi | 12 ++++--------
> >  1 file changed, 4 insertions(+), 8 deletions(-)
> > 
> > diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> > b/arch/arm64/boot/dts/freescale/imx8mp.dtsi index
> > cc406bb338fe..e0ca82ff6f15 100644
> > --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> > +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi
> > @@ -306,8 +306,7 @@ soc: soc@0 {
> > 
> >  		etm0: etm@28440000 {
> >  			compatible = "arm,coresight-etm4x", 
> "arm,primecell";
> > -			reg = <0x28440000 0x10000>;
> > -			arm,primecell-periphid = <0xbb95d>;
> > +			reg = <0x28440000 0x1000>;
> >  			cpu = <&A53_0>;
> >  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
> >  			clock-names = "apb_pclk";
> > @@ -323,8 +322,7 @@ etm0_out_port: endpoint {
> > 
> >  		etm1: etm@28540000 {
> >  			compatible = "arm,coresight-etm4x", 
> "arm,primecell";
> > -			reg = <0x28540000 0x10000>;
> > -			arm,primecell-periphid = <0xbb95d>;
> > +			reg = <0x28540000 0x1000>;
> >  			cpu = <&A53_1>;
> >  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
> >  			clock-names = "apb_pclk";
> > @@ -340,8 +338,7 @@ etm1_out_port: endpoint {
> > 
> >  		etm2: etm@28640000 {
> >  			compatible = "arm,coresight-etm4x", 
> "arm,primecell";
> > -			reg = <0x28640000 0x10000>;
> > -			arm,primecell-periphid = <0xbb95d>;
> > +			reg = <0x28640000 0x1000>;
> >  			cpu = <&A53_2>;
> >  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
> >  			clock-names = "apb_pclk";
> > @@ -357,8 +354,7 @@ etm2_out_port: endpoint {
> > 
> >  		etm3: etm@28740000 {
> >  			compatible = "arm,coresight-etm4x", 
> "arm,primecell";
> > -			reg = <0x28740000 0x10000>;
> > -			arm,primecell-periphid = <0xbb95d>;
> > +			reg = <0x28740000 0x1000>;
> >  			cpu = <&A53_3>;
> >  			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
> >  			clock-names = "apb_pclk";
> 
>
Robin Murphy July 6, 2023, 11:06 a.m. UTC | #3
On 2023-07-06 09:23, Lucas Stach wrote:
> Hi Alexander,
> 
> Am Donnerstag, dem 06.07.2023 um 07:06 +0200 schrieb Alexander Stein:
>> Hi Frank,
>>
>> Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
>>> The reg size of etm nodes is incorrectly set to 64k instead of 4k. This
>>> leads to a crash when calling amba_read_periphid().  After corrected reg
>>> size, amba_read_periphid() retrieve the correct periphid.
>>> arm,primecell-periphid were removed from the etm nodes.
>>
>> So this means the reference manual is wrong here? It clearly states the size
>> is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
>> On a side note: Is imx8mq affected by this as well? The DAP memory table lists
>> similar sizes in the RM .
>>
> Note that the 64K MMIO space per device is really an alignment thing.
> It's a recommendation from ARM to allow individual device MMIO regions
> to be mapped on kernels with 64K page size. Most of the time the real
> MMIO space occupied by the device is actually much smaller than 64K.

Indeed, it's quite common for TRM memory maps to be written in terms of 
the interconnect configuration, i.e. from the point of view of the 
interconnect itself, that whole range of address space is assigned to 
that peripheral, and it may even be true that the entire range is routed 
to the port where that peripheral is connected. However what's of more 
interest for DT is how much of that range the peripheral itself actually 
decodes.

Robin.

> 
> Regards,
> Lucas
> 
>> Best regards,
>> Alexander
>>
>>> Signed-off-by: Frank Li <Frank.Li@nxp.com>
>>> ---
>>>   arch/arm64/boot/dts/freescale/imx8mp.dtsi | 12 ++++--------
>>>   1 file changed, 4 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
>>> b/arch/arm64/boot/dts/freescale/imx8mp.dtsi index
>>> cc406bb338fe..e0ca82ff6f15 100644
>>> --- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
>>> +++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi
>>> @@ -306,8 +306,7 @@ soc: soc@0 {
>>>
>>>   		etm0: etm@28440000 {
>>>   			compatible = "arm,coresight-etm4x",
>> "arm,primecell";
>>> -			reg = <0x28440000 0x10000>;
>>> -			arm,primecell-periphid = <0xbb95d>;
>>> +			reg = <0x28440000 0x1000>;
>>>   			cpu = <&A53_0>;
>>>   			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>>>   			clock-names = "apb_pclk";
>>> @@ -323,8 +322,7 @@ etm0_out_port: endpoint {
>>>
>>>   		etm1: etm@28540000 {
>>>   			compatible = "arm,coresight-etm4x",
>> "arm,primecell";
>>> -			reg = <0x28540000 0x10000>;
>>> -			arm,primecell-periphid = <0xbb95d>;
>>> +			reg = <0x28540000 0x1000>;
>>>   			cpu = <&A53_1>;
>>>   			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>>>   			clock-names = "apb_pclk";
>>> @@ -340,8 +338,7 @@ etm1_out_port: endpoint {
>>>
>>>   		etm2: etm@28640000 {
>>>   			compatible = "arm,coresight-etm4x",
>> "arm,primecell";
>>> -			reg = <0x28640000 0x10000>;
>>> -			arm,primecell-periphid = <0xbb95d>;
>>> +			reg = <0x28640000 0x1000>;
>>>   			cpu = <&A53_2>;
>>>   			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>>>   			clock-names = "apb_pclk";
>>> @@ -357,8 +354,7 @@ etm2_out_port: endpoint {
>>>
>>>   		etm3: etm@28740000 {
>>>   			compatible = "arm,coresight-etm4x",
>> "arm,primecell";
>>> -			reg = <0x28740000 0x10000>;
>>> -			arm,primecell-periphid = <0xbb95d>;
>>> +			reg = <0x28740000 0x1000>;
>>>   			cpu = <&A53_3>;
>>>   			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
>>>   			clock-names = "apb_pclk";
>>
>>
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Frank Li July 6, 2023, 2:39 p.m. UTC | #4
On Thu, Jul 06, 2023 at 12:06:19PM +0100, Robin Murphy wrote:
> > > 
> > > Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> > > > The reg size of etm nodes is incorrectly set to 64k instead of 4k. This
> > > > leads to a crash when calling amba_read_periphid().  After corrected reg
> > > > size, amba_read_periphid() retrieve the correct periphid.
> > > > arm,primecell-periphid were removed from the etm nodes.
> > > 
> > > So this means the reference manual is wrong here? It clearly states the size
> > > is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
> > > On a side note: Is imx8mq affected by this as well? The DAP memory table lists
> > > similar sizes in the RM .
> > > 
> > Note that the 64K MMIO space per device is really an alignment thing.
> > It's a recommendation from ARM to allow individual device MMIO regions
> > to be mapped on kernels with 64K page size. Most of the time the real
> > MMIO space occupied by the device is actually much smaller than 64K.
> 
> Indeed, it's quite common for TRM memory maps to be written in terms of the
> interconnect configuration, i.e. from the point of view of the interconnect
> itself, that whole range of address space is assigned to that peripheral,
> and it may even be true that the entire range is routed to the port where
> that peripheral is connected. However what's of more interest for DT is how
> much of that range the peripheral itself actually decodes.

Yes, there are not problem by mapping bigger space in most case.

amba bus's periphal use close to end of region to show device's identical
information.

In drivers/amba/bus.c, 

amba_read_periphid()
{
	...
	size = resource_size(&dev->res);
	...
	for (pid = 0, i = 0; i < 4; i++)
		pid |= (readl(tmp + size - 0x20 + 4 * i) & 255) << (i * 8);
}

So the range in DTS for arm,primecell should be actual IP address space.

> 
> Robin.
> 
> > 
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel@lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Alexander Stein July 7, 2023, 5:34 a.m. UTC | #5
Hi Frank,

Am Donnerstag, 6. Juli 2023, 16:39:07 CEST schrieb Frank Li:
> On Thu, Jul 06, 2023 at 12:06:19PM +0100, Robin Murphy wrote:
> > > > Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> > > > > The reg size of etm nodes is incorrectly set to 64k instead of 4k.
> > > > > This
> > > > > leads to a crash when calling amba_read_periphid().  After corrected
> > > > > reg
> > > > > size, amba_read_periphid() retrieve the correct periphid.
> > > > > arm,primecell-periphid were removed from the etm nodes.
> > > > 
> > > > So this means the reference manual is wrong here? It clearly states
> > > > the size is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
> > > > On a side note: Is imx8mq affected by this as well? The DAP memory
> > > > table lists similar sizes in the RM .
> > > 
> > > Note that the 64K MMIO space per device is really an alignment thing.
> > > It's a recommendation from ARM to allow individual device MMIO regions
> > > to be mapped on kernels with 64K page size. Most of the time the real
> > > MMIO space occupied by the device is actually much smaller than 64K.
> > 
> > Indeed, it's quite common for TRM memory maps to be written in terms of
> > the
> > interconnect configuration, i.e. from the point of view of the
> > interconnect
> > itself, that whole range of address space is assigned to that peripheral,
> > and it may even be true that the entire range is routed to the port where
> > that peripheral is connected. However what's of more interest for DT is
> > how
> > much of that range the peripheral itself actually decodes.
> 
> Yes, there are not problem by mapping bigger space in most case.
> 
> amba bus's periphal use close to end of region to show device's identical
> information.

Ah, thanks for the explanation. This make things more clear.
But on the other is it sensible to assume the memory resource size to fit the 
IP address space? It appears to me the size is fixed to 4kiB anyway. Would it 
make more sense to read the values from the address "base + 4K - x" instead of 
"base + size - x"?

Best regards,
Alexander

> In drivers/amba/bus.c,
> 
> amba_read_periphid()
> {
> 	...
> 	size = resource_size(&dev->res);
> 	...
> 	for (pid = 0, i = 0; i < 4; i++)
> 		pid |= (readl(tmp + size - 0x20 + 4 * i) & 255) << (i * 
8);
> }
> 
> So the range in DTS for arm,primecell should be actual IP address space.
> 
> > Robin.
> > 
> > > _______________________________________________
> > > linux-arm-kernel mailing list
> > > linux-arm-kernel@lists.infradead.org
> > > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Robin Murphy July 7, 2023, 8:50 a.m. UTC | #6
On 2023-07-07 06:34, Alexander Stein wrote:
> Hi Frank,
> 
> Am Donnerstag, 6. Juli 2023, 16:39:07 CEST schrieb Frank Li:
>> On Thu, Jul 06, 2023 at 12:06:19PM +0100, Robin Murphy wrote:
>>>>> Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
>>>>>> The reg size of etm nodes is incorrectly set to 64k instead of 4k.
>>>>>> This
>>>>>> leads to a crash when calling amba_read_periphid().  After corrected
>>>>>> reg
>>>>>> size, amba_read_periphid() retrieve the correct periphid.
>>>>>> arm,primecell-periphid were removed from the etm nodes.
>>>>>
>>>>> So this means the reference manual is wrong here? It clearly states
>>>>> the size is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
>>>>> On a side note: Is imx8mq affected by this as well? The DAP memory
>>>>> table lists similar sizes in the RM .
>>>>
>>>> Note that the 64K MMIO space per device is really an alignment thing.
>>>> It's a recommendation from ARM to allow individual device MMIO regions
>>>> to be mapped on kernels with 64K page size. Most of the time the real
>>>> MMIO space occupied by the device is actually much smaller than 64K.
>>>
>>> Indeed, it's quite common for TRM memory maps to be written in terms of
>>> the
>>> interconnect configuration, i.e. from the point of view of the
>>> interconnect
>>> itself, that whole range of address space is assigned to that peripheral,
>>> and it may even be true that the entire range is routed to the port where
>>> that peripheral is connected. However what's of more interest for DT is
>>> how
>>> much of that range the peripheral itself actually decodes.
>>
>> Yes, there are not problem by mapping bigger space in most case.
>>
>> amba bus's periphal use close to end of region to show device's identical
>> information.
> 
> Ah, thanks for the explanation. This make things more clear.
> But on the other is it sensible to assume the memory resource size to fit the
> IP address space? It appears to me the size is fixed to 4kiB anyway. Would it
> make more sense to read the values from the address "base + 4K - x" instead of
> "base + size - x"?

The size of PrimeCell components in general isn't necessarily 4KB 
though, and the ID registers were defined relative to the *end* of the 
register space. The old PrimeCell standards evolved into the CoreSight 
spec, and from the oldest version of that I can easily link to[1]:

"Each component occupies one or more contiguous 4KB blocks of address 
space. Where a component occupies more than one 4KB block, these 
registers must appear in the highest 4KB block."

(FWIW the latest Coresight 3.0 spec relaxes this restriction, but we 
tend to model newer stuff as platform drivers with explicit DT/ACPI 
identifiers rather than amba drivers anyway)

Thanks,
Robin.

[1] https://developer.arm.com/documentation/ihi0029/d/?lang=en

> 
> Best regards,
> Alexander
> 
>> In drivers/amba/bus.c,
>>
>> amba_read_periphid()
>> {
>> 	...
>> 	size = resource_size(&dev->res);
>> 	...
>> 	for (pid = 0, i = 0; i < 4; i++)
>> 		pid |= (readl(tmp + size - 0x20 + 4 * i) & 255) << (i *
> 8);
>> }
>>
>> So the range in DTS for arm,primecell should be actual IP address space.
>>
>>> Robin.
>>>
>>>> _______________________________________________
>>>> linux-arm-kernel mailing list
>>>> linux-arm-kernel@lists.infradead.org
>>>> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> 
>
Lucas Stach July 7, 2023, 8:56 a.m. UTC | #7
Am Freitag, dem 07.07.2023 um 07:34 +0200 schrieb Alexander Stein:
> Hi Frank,
> 
> Am Donnerstag, 6. Juli 2023, 16:39:07 CEST schrieb Frank Li:
> > On Thu, Jul 06, 2023 at 12:06:19PM +0100, Robin Murphy wrote:
> > > > > Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> > > > > > The reg size of etm nodes is incorrectly set to 64k instead of 4k.
> > > > > > This
> > > > > > leads to a crash when calling amba_read_periphid().  After corrected
> > > > > > reg
> > > > > > size, amba_read_periphid() retrieve the correct periphid.
> > > > > > arm,primecell-periphid were removed from the etm nodes.
> > > > > 
> > > > > So this means the reference manual is wrong here? It clearly states
> > > > > the size is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
> > > > > On a side note: Is imx8mq affected by this as well? The DAP memory
> > > > > table lists similar sizes in the RM .
> > > > 
> > > > Note that the 64K MMIO space per device is really an alignment thing.
> > > > It's a recommendation from ARM to allow individual device MMIO regions
> > > > to be mapped on kernels with 64K page size. Most of the time the real
> > > > MMIO space occupied by the device is actually much smaller than 64K.
> > > 
> > > Indeed, it's quite common for TRM memory maps to be written in terms of
> > > the
> > > interconnect configuration, i.e. from the point of view of the
> > > interconnect
> > > itself, that whole range of address space is assigned to that peripheral,
> > > and it may even be true that the entire range is routed to the port where
> > > that peripheral is connected. However what's of more interest for DT is
> > > how
> > > much of that range the peripheral itself actually decodes.
> > 
> > Yes, there are not problem by mapping bigger space in most case.
> > 
> > amba bus's periphal use close to end of region to show device's identical
> > information.
> 
> Ah, thanks for the explanation. This make things more clear.
> But on the other is it sensible to assume the memory resource size to fit the 
> IP address space? It appears to me the size is fixed to 4kiB anyway. Would it 
> make more sense to read the values from the address "base + 4K - x" instead of 
> "base + size - x"?
> 
Huh? Why would AMBA peripherals be fixed to 4K in MMIO size? The ID
detection by reading offsets at the end of the MMIO range is generic
AMBA bus behavior.

The DT should declare the real peripheral MMIO size. Declaring the
aligned size is not much of a problem on 64bit systems with huge
address spaces but it still wastes vmap space when remapping the MMIO
range. When the real peripheral decode size is known it should always
be preferred over the aligned size.

Regards,
Lucas

> Best regards,
> Alexander
> 
> > In drivers/amba/bus.c,
> > 
> > amba_read_periphid()
> > {
> > 	...
> > 	size = resource_size(&dev->res);
> > 	...
> > 	for (pid = 0, i = 0; i < 4; i++)
> > 		pid |= (readl(tmp + size - 0x20 + 4 * i) & 255) << (i * 
> 8);
> > }
> > 
> > So the range in DTS for arm,primecell should be actual IP address space.
> > 
> > > Robin.
> > > 
> > > > _______________________________________________
> > > > linux-arm-kernel mailing list
> > > > linux-arm-kernel@lists.infradead.org
> > > > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> 
>
Alexander Stein July 7, 2023, 12:25 p.m. UTC | #8
Hi Robin,

Am Freitag, 7. Juli 2023, 10:50:31 CEST schrieb Robin Murphy:
> On 2023-07-07 06:34, Alexander Stein wrote:
> > Hi Frank,
> > 
> > Am Donnerstag, 6. Juli 2023, 16:39:07 CEST schrieb Frank Li:
> >> On Thu, Jul 06, 2023 at 12:06:19PM +0100, Robin Murphy wrote:
> >>>>> Am Mittwoch, 5. Juli 2023, 22:59:53 CEST schrieb Frank Li:
> >>>>>> The reg size of etm nodes is incorrectly set to 64k instead of 4k.
> >>>>>> This
> >>>>>> leads to a crash when calling amba_read_periphid().  After corrected
> >>>>>> reg
> >>>>>> size, amba_read_periphid() retrieve the correct periphid.
> >>>>>> arm,primecell-periphid were removed from the etm nodes.
> >>>>> 
> >>>>> So this means the reference manual is wrong here? It clearly states
> >>>>> the size is 64kiB. Reference Manual i.MX8MP Rev 1. 06/2021
> >>>>> On a side note: Is imx8mq affected by this as well? The DAP memory
> >>>>> table lists similar sizes in the RM .
> >>>> 
> >>>> Note that the 64K MMIO space per device is really an alignment thing.
> >>>> It's a recommendation from ARM to allow individual device MMIO regions
> >>>> to be mapped on kernels with 64K page size. Most of the time the real
> >>>> MMIO space occupied by the device is actually much smaller than 64K.
> >>> 
> >>> Indeed, it's quite common for TRM memory maps to be written in terms of
> >>> the
> >>> interconnect configuration, i.e. from the point of view of the
> >>> interconnect
> >>> itself, that whole range of address space is assigned to that
> >>> peripheral,
> >>> and it may even be true that the entire range is routed to the port
> >>> where
> >>> that peripheral is connected. However what's of more interest for DT is
> >>> how
> >>> much of that range the peripheral itself actually decodes.
> >> 
> >> Yes, there are not problem by mapping bigger space in most case.
> >> 
> >> amba bus's periphal use close to end of region to show device's identical
> >> information.
> > 
> > Ah, thanks for the explanation. This make things more clear.
> > But on the other is it sensible to assume the memory resource size to fit
> > the IP address space? It appears to me the size is fixed to 4kiB anyway.
> > Would it make more sense to read the values from the address "base + 4K -
> > x" instead of "base + size - x"?
> 
> The size of PrimeCell components in general isn't necessarily 4KB
> though, and the ID registers were defined relative to the *end* of the
> register space. The old PrimeCell standards evolved into the CoreSight
> spec, and from the oldest version of that I can easily link to[1]:
> 
> "Each component occupies one or more contiguous 4KB blocks of address
> space. Where a component occupies more than one 4KB block, these
> registers must appear in the highest 4KB block."
> 
> (FWIW the latest Coresight 3.0 spec relaxes this restriction, but we
> tend to model newer stuff as platform drivers with explicit DT/ACPI
> identifiers rather than amba drivers anyway)

Ah, I wasn't aware the register space for PrimeCells/CoreSight could be larger 
than 4k. So the exact size must be known and used in DT. Thanks for 
explanation.

Best regards,
Alexander

> Thanks,
> Robin.
> 
> [1] https://developer.arm.com/documentation/ihi0029/d/?lang=en
> 
> > Best regards,
> > Alexander
> > 
> >> In drivers/amba/bus.c,
> >> 
> >> amba_read_periphid()
> >> {
> >> 
> >> 	...
> >> 	size = resource_size(&dev->res);
> >> 	...
> >> 	for (pid = 0, i = 0; i < 4; i++)
> >> 	
> >> 		pid |= (readl(tmp + size - 0x20 + 4 * i) & 255) << (i *
> > 
> > 8);
> > 
> >> }
> >> 
> >> So the range in DTS for arm,primecell should be actual IP address space.
> >> 
> >>> Robin.
> >>> 
> >>>> _______________________________________________
> >>>> linux-arm-kernel mailing list
> >>>> linux-arm-kernel@lists.infradead.org
> >>>> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
diff mbox series

Patch

diff --git a/arch/arm64/boot/dts/freescale/imx8mp.dtsi b/arch/arm64/boot/dts/freescale/imx8mp.dtsi
index cc406bb338fe..e0ca82ff6f15 100644
--- a/arch/arm64/boot/dts/freescale/imx8mp.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mp.dtsi
@@ -306,8 +306,7 @@  soc: soc@0 {
 
 		etm0: etm@28440000 {
 			compatible = "arm,coresight-etm4x", "arm,primecell";
-			reg = <0x28440000 0x10000>;
-			arm,primecell-periphid = <0xbb95d>;
+			reg = <0x28440000 0x1000>;
 			cpu = <&A53_0>;
 			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
 			clock-names = "apb_pclk";
@@ -323,8 +322,7 @@  etm0_out_port: endpoint {
 
 		etm1: etm@28540000 {
 			compatible = "arm,coresight-etm4x", "arm,primecell";
-			reg = <0x28540000 0x10000>;
-			arm,primecell-periphid = <0xbb95d>;
+			reg = <0x28540000 0x1000>;
 			cpu = <&A53_1>;
 			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
 			clock-names = "apb_pclk";
@@ -340,8 +338,7 @@  etm1_out_port: endpoint {
 
 		etm2: etm@28640000 {
 			compatible = "arm,coresight-etm4x", "arm,primecell";
-			reg = <0x28640000 0x10000>;
-			arm,primecell-periphid = <0xbb95d>;
+			reg = <0x28640000 0x1000>;
 			cpu = <&A53_2>;
 			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
 			clock-names = "apb_pclk";
@@ -357,8 +354,7 @@  etm2_out_port: endpoint {
 
 		etm3: etm@28740000 {
 			compatible = "arm,coresight-etm4x", "arm,primecell";
-			reg = <0x28740000 0x10000>;
-			arm,primecell-periphid = <0xbb95d>;
+			reg = <0x28740000 0x1000>;
 			cpu = <&A53_3>;
 			clocks = <&clk IMX8MP_CLK_MAIN_AXI>;
 			clock-names = "apb_pclk";