[PATCH mmotm] x86/numa: fix build when CONFIG_ACPI is not set

[PATCH mmotm] x86/numa: fix build when CONFIG_ACPI is not set

[Randy Dunlap]

From: Randy Dunlap <rdunlap@infradead.org>

Fix build errors when CONFIG_ACPI is not set/enabled by adding
<acpi/acpi_numa.h> in the #else (!CONFIG_ACPI) block.

../arch/x86/mm/numa.c: In function ‘numa_setup’:
../arch/x86/mm/numa.c:43:3: error: implicit declaration of function ‘disable_srat’; did you mean ‘disable_irq’? [-Werror=implicit-function-declaration]
   disable_srat();
   ^~~~~~~~~~~~
../arch/x86/mm/numa.c:45:3: error: implicit declaration of function ‘disable_hmat’; did you mean ‘disable_dma’? [-Werror=implicit-function-declaration]
   disable_hmat();
   ^~~~~~~~~~~~


Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Dan Williams <dan.j.williams@intel.com>
---
 include/linux/acpi.h |    2 ++
 1 file changed, 2 insertions(+)

--- mmotm-2020-0824-1606.orig/include/linux/acpi.h
+++ mmotm-2020-0824-1606/include/linux/acpi.h
@@ -709,6 +709,8 @@ static inline u64 acpi_arch_get_root_poi
 #define ACPI_HANDLE_FWNODE(fwnode)	(NULL)
 #define ACPI_DEVICE_CLASS(_cls, _msk)	.cls = (0), .cls_msk = (0),
 
+#include <acpi/acpi_numa.h>
+
 struct fwnode_handle;
 
 static inline bool acpi_dev_found(const char *hid)

Re: [PATCH mmotm] x86/numa: fix build when CONFIG_ACPI is not set

[Andrew Morton]

On Tue, 25 Aug 2020 08:20:27 -0700 Randy Dunlap <rdunlap@infradead.org> wrote:

> From: Randy Dunlap <rdunlap@infradead.org>
> 
> Fix build errors when CONFIG_ACPI is not set/enabled by adding
> <acpi/acpi_numa.h> in the #else (!CONFIG_ACPI) block.
> 
> ../arch/x86/mm/numa.c: In function ‘numa_setup’:
> ../arch/x86/mm/numa.c:43:3: error: implicit declaration of function ‘disable_srat’; did you mean ‘disable_irq’? [-Werror=implicit-function-declaration]
>    disable_srat();
>    ^~~~~~~~~~~~
> ../arch/x86/mm/numa.c:45:3: error: implicit declaration of function ‘disable_hmat’; did you mean ‘disable_dma’? [-Werror=implicit-function-declaration]
>    disable_hmat();
>    ^~~~~~~~~~~~
>
> ...
>
> --- mmotm-2020-0824-1606.orig/include/linux/acpi.h
> +++ mmotm-2020-0824-1606/include/linux/acpi.h
> @@ -709,6 +709,8 @@ static inline u64 acpi_arch_get_root_poi
>  #define ACPI_HANDLE_FWNODE(fwnode)	(NULL)
>  #define ACPI_DEVICE_CLASS(_cls, _msk)	.cls = (0), .cls_msk = (0),
>  
> +#include <acpi/acpi_numa.h>
> +
>  struct fwnode_handle;
>  
>  static inline bool acpi_dev_found(const char *hid)

I don't think there's anything in -next that caused this?  It's in
mainline also?

Re: [PATCH mmotm] x86/numa: fix build when CONFIG_ACPI is not set

[Randy Dunlap]

On 8/25/20 4:52 PM, Andrew Morton wrote:
> On Tue, 25 Aug 2020 08:20:27 -0700 Randy Dunlap <rdunlap@infradead.org> wrote:
> 
>> From: Randy Dunlap <rdunlap@infradead.org>
>>
>> Fix build errors when CONFIG_ACPI is not set/enabled by adding
>> <acpi/acpi_numa.h> in the #else (!CONFIG_ACPI) block.
>>
>> ../arch/x86/mm/numa.c: In function ‘numa_setup’:
>> ../arch/x86/mm/numa.c:43:3: error: implicit declaration of function ‘disable_srat’; did you mean ‘disable_irq’? [-Werror=implicit-function-declaration]
>>    disable_srat();
>>    ^~~~~~~~~~~~
>> ../arch/x86/mm/numa.c:45:3: error: implicit declaration of function ‘disable_hmat’; did you mean ‘disable_dma’? [-Werror=implicit-function-declaration]
>>    disable_hmat();
>>    ^~~~~~~~~~~~
>>
>> ...
>>
>> --- mmotm-2020-0824-1606.orig/include/linux/acpi.h
>> +++ mmotm-2020-0824-1606/include/linux/acpi.h
>> @@ -709,6 +709,8 @@ static inline u64 acpi_arch_get_root_poi
>>  #define ACPI_HANDLE_FWNODE(fwnode)	(NULL)
>>  #define ACPI_DEVICE_CLASS(_cls, _msk)	.cls = (0), .cls_msk = (0),
>>  
>> +#include <acpi/acpi_numa.h>
>> +
>>  struct fwnode_handle;
>>  
>>  static inline bool acpi_dev_found(const char *hid)
> 
> I don't think there's anything in -next that caused this?  It's in
> mainline also?
> 

Calls to disable_srat() and disable_hmat() are not in v5.9-rc2.
If they are in mainline, they were added after -rc2.

See these patches in mmotm:
x86-numa-add-nohmat-option.patch
x86-numa-cleanup-configuration-dependent-command-line-options.patch


-- 
~Randy