[PATCH v1 -next 0/2] RISC-V ACPI: Assorted fixes

[PATCH v1 -next 0/2] RISC-V ACPI: Assorted fixes

[Sunil V L]

This series fixes couple of issues required to complete the previously
merged basic ACPI support series.

1) Kernel test bot has found a new warning using the sparse tool. Though
I believe this warning exists for other architectures also,
I don't see an issue fixing this for RISC-V.

2) When sscofpmf extension is enabled, the perf driver tries to
find out the root INTC domain to register the over flow interrupt
handler. However, this function is DT based and needs to be enhanced
to work on ACPI based systems.

Sunil V L (2):
  RISC-V: ACPI : Fix for usage of pointers in different address space
  RISC-V/perf: Use standard interface to get INTC domain

 arch/riscv/kernel/acpi.c     |  4 ++--
 drivers/perf/riscv_pmu_sbi.c | 17 ++---------------
 2 files changed, 4 insertions(+), 17 deletions(-)

-- 
2.34.1

[PATCH v1 -next 1/2] RISC-V: ACPI : Fix for usage of pointers in different address space

[Sunil V L]

The arch specific __acpi_map_table can be wrapper around either
early_memremap or early_ioremap. But early_memremap
routine works with normal pointers whereas __acpi_map_table expects
pointers in iomem address space. This causes kernel test bot to fail
while using the sparse tool. Fix the issue by using early_ioremap and
similar fix done for __acpi_unmap_table.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202305201427.I7QhPjNW-lkp@intel.com/
---
 arch/riscv/kernel/acpi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/riscv/kernel/acpi.c b/arch/riscv/kernel/acpi.c
index df5a45a2eb93..5ee03ebab80e 100644
--- a/arch/riscv/kernel/acpi.c
+++ b/arch/riscv/kernel/acpi.c
@@ -204,7 +204,7 @@ void __init __iomem *__acpi_map_table(unsigned long phys, unsigned long size)
 	if (!size)
 		return NULL;
 
-	return early_memremap(phys, size);
+	return early_ioremap(phys, size);
 }
 
 void __init __acpi_unmap_table(void __iomem *map, unsigned long size)
@@ -212,7 +212,7 @@ void __init __acpi_unmap_table(void __iomem *map, unsigned long size)
 	if (!map || !size)
 		return;
 
-	early_memunmap(map, size);
+	early_iounmap(map, size);
 }
 
 void *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
-- 
2.34.1

Re: [PATCH v1 -next 1/2] RISC-V: ACPI : Fix for usage of pointers in different address space

[Conor Dooley]

[-- Attachment #1: Type: text/plain, Size: 1708 bytes --]

Hey Sunil,

On Mon, Jun 05, 2023 at 08:05:11PM +0530, Sunil V L wrote:
> The arch specific __acpi_map_table can be wrapper around either
> early_memremap or early_ioremap. But early_memremap
> routine works with normal pointers whereas __acpi_map_table expects
> pointers in iomem address space. This causes kernel test bot to fail
> while using the sparse tool. Fix the issue by using early_ioremap and
> similar fix done for __acpi_unmap_table.
> 
> Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202305201427.I7QhPjNW-lkp@intel.com/

Is this
Fixes: a91a9ffbd3a5 ("RISC-V: Add support to build the ACPI core")
?
Otherwise,
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>

Thanks,
Conor.

> ---
>  arch/riscv/kernel/acpi.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/riscv/kernel/acpi.c b/arch/riscv/kernel/acpi.c
> index df5a45a2eb93..5ee03ebab80e 100644
> --- a/arch/riscv/kernel/acpi.c
> +++ b/arch/riscv/kernel/acpi.c
> @@ -204,7 +204,7 @@ void __init __iomem *__acpi_map_table(unsigned long phys, unsigned long size)
>  	if (!size)
>  		return NULL;
>  
> -	return early_memremap(phys, size);
> +	return early_ioremap(phys, size);
>  }
>  
>  void __init __acpi_unmap_table(void __iomem *map, unsigned long size)
> @@ -212,7 +212,7 @@ void __init __acpi_unmap_table(void __iomem *map, unsigned long size)
>  	if (!map || !size)
>  		return;
>  
> -	early_memunmap(map, size);
> +	early_iounmap(map, size);
>  }
>  
>  void *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
> -- 
> 2.34.1
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

Re: [PATCH v1 -next 1/2] RISC-V: ACPI : Fix for usage of pointers in different address space

[Sunil V L]

Hi Conor,

On Tue, Jun 06, 2023 at 04:35:29PM +0100, Conor Dooley wrote:
> Hey Sunil,
> 
> On Mon, Jun 05, 2023 at 08:05:11PM +0530, Sunil V L wrote:
> > The arch specific __acpi_map_table can be wrapper around either
> > early_memremap or early_ioremap. But early_memremap
> > routine works with normal pointers whereas __acpi_map_table expects
> > pointers in iomem address space. This causes kernel test bot to fail
> > while using the sparse tool. Fix the issue by using early_ioremap and
> > similar fix done for __acpi_unmap_table.
> > 
> > Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202305201427.I7QhPjNW-lkp@intel.com/
> 
> Is this
> Fixes: a91a9ffbd3a5 ("RISC-V: Add support to build the ACPI core")
> ?
Yes, I missed it. Will add.

Thanks!
Sunil

[PATCH v1 -next 2/2] RISC-V/perf: Use standard interface to get INTC domain

[Sunil V L]

Currently the PMU driver is using DT based lookup to
find the INTC node for sscofpmf extension. This will not work
for ACPI based systems causing the driver to fail to register
the PMU overflow interrupt handler.

Hence, change the code to use the standard interface to find
the INTC node which works irrespective of DT or ACPI.

Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
---
 drivers/perf/riscv_pmu_sbi.c | 17 ++---------------
 1 file changed, 2 insertions(+), 15 deletions(-)

diff --git a/drivers/perf/riscv_pmu_sbi.c b/drivers/perf/riscv_pmu_sbi.c
index 4f3ac296b3e2..0bc491252a44 100644
--- a/drivers/perf/riscv_pmu_sbi.c
+++ b/drivers/perf/riscv_pmu_sbi.c
@@ -739,7 +739,6 @@ static int pmu_sbi_setup_irqs(struct riscv_pmu *pmu, struct platform_device *pde
 {
 	int ret;
 	struct cpu_hw_events __percpu *hw_events = pmu->hw_events;
-	struct device_node *cpu, *child;
 	struct irq_domain *domain = NULL;
 
 	if (riscv_isa_extension_available(NULL, SSCOFPMF)) {
@@ -756,20 +755,8 @@ static int pmu_sbi_setup_irqs(struct riscv_pmu *pmu, struct platform_device *pde
 	if (!riscv_pmu_use_irq)
 		return -EOPNOTSUPP;
 
-	for_each_of_cpu_node(cpu) {
-		child = of_get_compatible_child(cpu, "riscv,cpu-intc");
-		if (!child) {
-			pr_err("Failed to find INTC node\n");
-			of_node_put(cpu);
-			return -ENODEV;
-		}
-		domain = irq_find_host(child);
-		of_node_put(child);
-		if (domain) {
-			of_node_put(cpu);
-			break;
-		}
-	}
+	domain = irq_find_matching_fwnode(riscv_get_intc_hwnode(),
+					  DOMAIN_BUS_ANY);
 	if (!domain) {
 		pr_err("Failed to find INTC IRQ root domain\n");
 		return -ENODEV;
-- 
2.34.1

Re: [PATCH v1 -next 2/2] RISC-V/perf: Use standard interface to get INTC domain

[Conor Dooley]

[-- Attachment #1: Type: text/plain, Size: 1997 bytes --]

On Mon, Jun 05, 2023 at 08:05:12PM +0530, Sunil V L wrote:
> Currently the PMU driver is using DT based lookup to
> find the INTC node for sscofpmf extension. This will not work
> for ACPI based systems causing the driver to fail to register
> the PMU overflow interrupt handler.
> 
> Hence, change the code to use the standard interface to find
> the INTC node which works irrespective of DT or ACPI.
> 
> Signed-off-by: Sunil V L <sunilvl@ventanamicro.com>
> ---
>  drivers/perf/riscv_pmu_sbi.c | 17 ++---------------
>  1 file changed, 2 insertions(+), 15 deletions(-)
> 
> diff --git a/drivers/perf/riscv_pmu_sbi.c b/drivers/perf/riscv_pmu_sbi.c
> index 4f3ac296b3e2..0bc491252a44 100644
> --- a/drivers/perf/riscv_pmu_sbi.c
> +++ b/drivers/perf/riscv_pmu_sbi.c
> @@ -739,7 +739,6 @@ static int pmu_sbi_setup_irqs(struct riscv_pmu *pmu, struct platform_device *pde
>  {
>  	int ret;
>  	struct cpu_hw_events __percpu *hw_events = pmu->hw_events;
> -	struct device_node *cpu, *child;
>  	struct irq_domain *domain = NULL;
>  
>  	if (riscv_isa_extension_available(NULL, SSCOFPMF)) {
> @@ -756,20 +755,8 @@ static int pmu_sbi_setup_irqs(struct riscv_pmu *pmu, struct platform_device *pde
>  	if (!riscv_pmu_use_irq)
>  		return -EOPNOTSUPP;
>  
> -	for_each_of_cpu_node(cpu) {
> -		child = of_get_compatible_child(cpu, "riscv,cpu-intc");
> -		if (!child) {
> -			pr_err("Failed to find INTC node\n");
> -			of_node_put(cpu);
> -			return -ENODEV;
> -		}
> -		domain = irq_find_host(child);
> -		of_node_put(child);
> -		if (domain) {
> -			of_node_put(cpu);
> -			break;
> -		}
> -	}
> +	domain = irq_find_matching_fwnode(riscv_get_intc_hwnode(),
> +					  DOMAIN_BUS_ANY);

This riscv_get_intc_hwnode stuff is horrific, but that's besides the
point.
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>

Cheers,
Conor.

>  	if (!domain) {
>  		pr_err("Failed to find INTC IRQ root domain\n");
>  		return -ENODEV;
> -- 
> 2.34.1
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]