RISC-V: ACPI improvements
Subsystems (3)
Sunil V L
maintainerv3 cover RISC-V: ACPI improvements
This series is a set of patches which were originally part of RFC v1 series
[1] to add ACPI support in RISC-V interrupt controllers. Since these patches are independent of the interrupt controllers, creating this new series which helps to merge instead of waiting for big series.
This set of patches primarily adds support below ECR [2] which is approved by the ASWG and adds below features.
- Get CBO block sizes from RHCT on ACPI based systems.
Additionally, the series contains a patch to improve acpi_os_ioremap().
[1] - https://lore.kernel.org/lkml/[email protected]/
[2] - https://drive.google.com/file/d/1sKbOa8m1UZw1JkquZYe3F1zQBN1xXsaf/view?usp=sharing
Changes since v2:
1) Modified acpi_get_cbo_block_size() not to take cpu parameter but follow same pattern as DT (Feedback from Samuel and Drew)
2) Dropped timer patch from the series since it is already applied.
3) Selected ARCH_KEEP_MEMBLOCK only if ACPI (Feedback from Alex).
4) Added RB tags received so far except RHCT patch which has changed quite significantly from previous version.
Changes since RFC v1:
1) Separated the patches from interrupt controller support series.
2) Addressed feedback from Andy and Drew.
3) Rebased to Palmer's for-next tree.
4) Added RB tags received on RFC v1.
Sunil V L (3):
RISC-V: ACPI: Enhance acpi_os_ioremap with MMIO remapping
RISC-V: ACPI: RHCT: Add function to get CBO block sizes
RISC-V: cacheflush: Initialize CBO variables on ACPI systems
arch/riscv/Kconfig | 1 + arch/riscv/include/asm/acpi.h | 6 +++ arch/riscv/kernel/acpi.c | 87 +++++++++++++++++++++++++++++++- arch/riscv/mm/cacheflush.c | 25 +++++++--- drivers/acpi/riscv/rhct.c | 93 +++++++++++++++++++++++++++++++++++ 5 files changed, 204 insertions(+), 8 deletions(-)
Signature · 1 line
2.39.2
Sunil V L
maintainerPATCH v3 1/3 RISC-V: ACPI: Enhance acpi_os_ioremap with MMIO remapping
Enhance the acpi_os_ioremap() to support opregions in MMIO space. Also, have strict checks using EFI memory map to allow remapping the RAM similar to arm64.
--- arch/riscv/Kconfig | 1 + arch/riscv/kernel/acpi.c | 87 +++++++++++++++++++++++++++++++++++++++- 2 files changed, 86 insertions(+), 2 deletions(-)
arch/riscv/Kconfig +1 -0
@@ -39,6 +39,7 @@ config RISCV select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST select ARCH_HAS_UBSAN_SANITIZE_ALL select ARCH_HAS_VDSO_DATA + select ARCH_KEEP_MEMBLOCK if ACPI select ARCH_OPTIONAL_KERNEL_RWX if ARCH_HAS_STRICT_KERNEL_RWX select ARCH_OPTIONAL_KERNEL_RWX_DEFAULT select ARCH_STACKWALK
arch/riscv/kernel/acpi.c +85 -2
101 lines not on this page · raw message
Signature · 1 line
2.39.2
Sunil V L
maintainerPATCH v3 2/3 RISC-V: ACPI: RHCT: Add function to get CBO block sizes
In reply to Sunil V LCache Block Operation (CBO) related block size in ACPI is provided by RHCT. Add support to read the CMO node in RHCT to get this information.
--- arch/riscv/include/asm/acpi.h | 6 +++ drivers/acpi/riscv/rhct.c | 93 +++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+)
arch/riscv/include/asm/acpi.h +6 -0
@@ -66,6 +66,8 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const char **isa);static inline int acpi_numa_get_nid (unsigned int cpu) {return NUMA_NO_NODE ; } +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size);#else static inline void acpi_init_rintc_map (void ) { }static inline struct acpi_madt_rintc *acpi_cpu_get_madt_rintc (int cpu)
@@ -79,6 +81,10 @@ static inline int acpi_get_riscv_isa(struct acpi_table_header *table,return -EINVAL ; } +static inline void acpi_get_cbo_block_size (struct acpi_table_header *table, +u32 *cbom_size,u32 *cboz_size, +u32 *cbop_size) { } +#endif /* CONFIG_ACPI */ #endif /*_ASM_ACPI_H*/
drivers/acpi/riscv/rhct.c +93 -0
102 lines not on this page · raw message
Signature · 1 line
2.39.2
Re: [PATCH v3 -next 2/3] RISC-V: ACPI: RHCT: Add function to get CBO block sizes
Quoted 62 lines · Sunil V L
Cache Block Operation (CBO) related block size in ACPI is provided by RHCT. Add support to read the CMO node in RHCT to get this information.
Signed-off-by Sunil V L <[email protected]>--- arch/riscv/include/asm/acpi.h | 6 +++ drivers/acpi/riscv/rhct.c | 93 +++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+)arch/riscv/include/asm/acpi.h +6
-0@@ -66,6 +66,8 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const char **isa);static inline int acpi_numa_get_nid (unsigned int cpu) {return NUMA_NO_NODE ; } +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size);#else static inline void acpi_init_rintc_map (void ) { }static inline struct acpi_madt_rintc *acpi_cpu_get_madt_rintc (int cpu)@@ -79,6 +81,10 @@ static inline int acpi_get_riscv_isa(struct acpi_table_header *table,return -EINVAL ; } +static inline void acpi_get_cbo_block_size (struct acpi_table_header *table, +u32 *cbom_size,u32 *cboz_size, +u32 *cbop_size) { } +#endif /* CONFIG_ACPI */ #endif /*_ASM_ACPI_H*/ drivers/acpi/riscv/rhct.c +14
-0@@ -8,6 +8,7 @@#define pr_fmt (fmt) "ACPI: RHCT: " fmt#include <linux/acpi.h> +#include <linux/bits.h> static struct acpi_table_header *acpi_get_rhct (void ) {@@ -81,3 +82,95 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, constreturn -1 ; } + +static void acpi_parse_hart_info_cmo_node (struct acpi_table_rhct *rhct, +struct acpi_rhct_hart_info *hart_info, +u32 *cbom_size,u32 *cboz_size,u32 *cbop_size) +{ +u32 size_hartinfo =sizeof (struct acpi_rhct_hart_info ); +u32 size_hdr =sizeof (struct acpi_rhct_node_header ); +struct acpi_rhct_node_header *ref_node; +struct acpi_rhct_cmo_node *cmo_node; +u32 *hart_info_node_offset; + + hart_info_node_offset =ACPI_ADD_PTR (u32, hart_info, size_hartinfo); +for (int i =0 ; i < hart_info->num_offsets; i++) {
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 107
Quoted 10 lines
drivers/acpi/riscv/rhct.c +10
-0+ ref_node = ACPI_ADD_PTR(struct acpi_rhct_node_header, + rhct, hart_info_node_offset[i]); +if (ref_node->type ==ACPI_RHCT_NODE_TYPE_CMO ) { + cmo_node = ACPI_ADD_PTR(struct acpi_rhct_cmo_node, + ref_node, size_hdr); +if (cbom_size && cmo_node->cbom_size <=30 ) { +if (!*cbom_size) { + *cbom_size =BIT (cmo_node->cbom_size); + }else if (*cbom_size != +BIT (cmo_node->cbom_size)) {
No need to break the if line, we can go to 100 chars. And then, since both the if and else if arms only have single statements, all the {} can be dropped too. Same comment for cboz and cbop.
Quoted 18 lines
drivers/acpi/riscv/rhct.c +18
-0+pr_warn ("CBOM size is not the same across harts\n" ); + } + } + +if (cboz_size && cmo_node->cboz_size <=30 ) { +if (!*cboz_size) { + *cboz_size =BIT (cmo_node->cboz_size); + }else if (*cboz_size != +BIT (cmo_node->cboz_size)) { +pr_warn ("CBOZ size is not the same across harts\n" ); + } + } + +if (cbop_size && cmo_node->cbop_size <=30 ) { +if (!*cbop_size) { + *cbop_size =BIT (cmo_node->cbop_size); + }else if (*cbop_size != +BIT (cmo_node->cbop_size)) {
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 135
Quoted 10 lines
drivers/acpi/riscv/rhct.c +10
-0+pr_warn ("CBOP size is not the same across harts\n" ); + } + } + } + } +} + +/* +* During early boot, the caller should call acpi_get_table() and pass its pointer to +* these functions(and free up later). At run time, since this table can be used
^ add a space here
Quoted 7 lines
drivers/acpi/riscv/rhct.c +7
-0+* multiple times, pass NULL so that the table remains in memory +*/ +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size) +{ +u32 size_hdr =sizeof (struct acpi_rhct_node_header ); +struct acpi_rhct_node_header *node, *end;
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 152
Quoted 10 lines
drivers/acpi/riscv/rhct.c +10
-0+struct acpi_rhct_hart_info *hart_info; +struct acpi_table_rhct *rhct; + +if (acpi_disabled) +return ; + +if (table) { + rhct = (struct acpi_table_rhct *)table; + }else { + rhct = (struct acpi_table_rhct *)acpi_get_rhct ();
Not an issue of this patch, but it seems like acpi_get_rhct() should return a struct acpi_table_rhct pointer instead of a struct acpi_table_header pointer since it's specifically returning an RHCT.
Quoted 27 lines
drivers/acpi/riscv/rhct.c +24
-0+if (!rhct) +return ; + } + +if (cbom_size) + *cbom_size =0 ; + +if (cboz_size) + *cboz_size =0 ; + +if (cbop_size) + *cbop_size =0 ; + + end = ACPI_ADD_PTR(struct acpi_rhct_node_header, rhct, rhct->header.length); + for (node = ACPI_ADD_PTR(struct acpi_rhct_node_header, rhct, rhct->node_offset); + node < end; + node = ACPI_ADD_PTR(struct acpi_rhct_node_header, node, node->length)) { +if (node->type ==ACPI_RHCT_NODE_TYPE_HART_INFO ) { + hart_info = ACPI_ADD_PTR(struct acpi_rhct_hart_info, node, size_hdr); +acpi_parse_hart_info_cmo_node (rhct, hart_info, cbom_size, + cboz_size, cbop_size); + } + } +}Signature · 1 line
2.39.2
Other than the nits
Thanks,
drew
Sunil V L
maintainerRe: [PATCH v3 -next 2/3] RISC-V: ACPI: RHCT: Add function to get CBO block sizes
Quoted 69 lines · Andrew Jones
Quoted 68 lines · Sunil V L
Cache Block Operation (CBO) related block size in ACPI is provided by RHCT. Add support to read the CMO node in RHCT to get this information.
Signed-off-by Sunil V L <[email protected]>--- arch/riscv/include/asm/acpi.h | 6 +++ drivers/acpi/riscv/rhct.c | 93 +++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+)arch/riscv/include/asm/acpi.h +6
-0@@ -66,6 +66,8 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const char **isa);static inline int acpi_numa_get_nid (unsigned int cpu) {return NUMA_NO_NODE ; } +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size);#else static inline void acpi_init_rintc_map (void ) { }static inline struct acpi_madt_rintc *acpi_cpu_get_madt_rintc (int cpu)@@ -79,6 +81,10 @@ static inline int acpi_get_riscv_isa(struct acpi_table_header *table,return -EINVAL ; } +static inline void acpi_get_cbo_block_size (struct acpi_table_header *table, +u32 *cbom_size,u32 *cboz_size, +u32 *cbop_size) { } +#endif /* CONFIG_ACPI */ #endif /*_ASM_ACPI_H*/ drivers/acpi/riscv/rhct.c +20
-0@@ -8,6 +8,7 @@#define pr_fmt (fmt) "ACPI: RHCT: " fmt#include <linux/acpi.h> +#include <linux/bits.h> static struct acpi_table_header *acpi_get_rhct (void ) {@@ -81,3 +82,95 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, constreturn -1 ; } + +static void acpi_parse_hart_info_cmo_node (struct acpi_table_rhct *rhct, +struct acpi_rhct_hart_info *hart_info, +u32 *cbom_size,u32 *cboz_size,u32 *cbop_size) +{ +u32 size_hartinfo =sizeof (struct acpi_rhct_hart_info ); +u32 size_hdr =sizeof (struct acpi_rhct_node_header ); +struct acpi_rhct_node_header *ref_node; +struct acpi_rhct_cmo_node *cmo_node; +u32 *hart_info_node_offset; + + hart_info_node_offset =ACPI_ADD_PTR (u32, hart_info, size_hartinfo); +for (int i =0 ; i < hart_info->num_offsets; i++) { + ref_node = ACPI_ADD_PTR(struct acpi_rhct_node_header, + rhct, hart_info_node_offset[i]); +if (ref_node->type ==ACPI_RHCT_NODE_TYPE_CMO ) { + cmo_node = ACPI_ADD_PTR(struct acpi_rhct_cmo_node, + ref_node, size_hdr); +if (cbom_size && cmo_node->cbom_size <=30 ) {
Quoted 9 lines
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 107
Quoted 4 lines
drivers/acpi/riscv/rhct.c +4
-0+if (!*cbom_size) { + *cbom_size =BIT (cmo_node->cbom_size); + }else if (*cbom_size != +BIT (cmo_node->cbom_size)) {No need to break the if line, we can go to 100 chars. And then, since both the if and else if arms only have single statements, all the {} can be dropped too. Same comment for cboz and cbop.
Yeah, it is a side effect of working on repos with different coding standards. It is interesting that checkpatch didn't recommend to remove the braces. Let me fix it in next revision. Thanks!.
Quoted 43 lines
Quoted 18 lines
drivers/acpi/riscv/rhct.c +18
-0+pr_warn ("CBOM size is not the same across harts\n" ); + } + } + +if (cboz_size && cmo_node->cboz_size <=30 ) { +if (!*cboz_size) { + *cboz_size =BIT (cmo_node->cboz_size); + }else if (*cboz_size != +BIT (cmo_node->cboz_size)) { +pr_warn ("CBOZ size is not the same across harts\n" ); + } + } + +if (cbop_size && cmo_node->cbop_size <=30 ) { +if (!*cbop_size) { + *cbop_size =BIT (cmo_node->cbop_size); + }else if (*cbop_size != +BIT (cmo_node->cbop_size)) {drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 135
Quoted 10 lines
drivers/acpi/riscv/rhct.c +10
-0+pr_warn ("CBOP size is not the same across harts\n" ); + } + } + } + } +} + +/* +* During early boot, the caller should call acpi_get_table() and pass its pointer to +* these functions(and free up later). At run time, since this table can be used ^ add a space here
Quoted 13 lines
drivers/acpi/riscv/rhct.c +13
-0+* multiple times, pass NULL so that the table remains in memory +*/ +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size) +{ +u32 size_hdr =sizeof (struct acpi_rhct_node_header ); +struct acpi_rhct_node_header *node, *end; +struct acpi_rhct_hart_info *hart_info; +struct acpi_table_rhct *rhct; + +if (acpi_disabled) +return ; +
Quoted 9 lines
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 152
Quoted 4 lines
drivers/acpi/riscv/rhct.c +4
-0+if (table) { + rhct = (struct acpi_table_rhct *)table; + }else { + rhct = (struct acpi_table_rhct *)acpi_get_rhct ();Not an issue of this patch, but it seems like
acpi_get_rhct()should return a structacpi_table_rhctpointer instead of a structacpi_table_headerpointer since it's specifically returning an RHCT.
Makes sense. Let me add a patch to improve this.
Quoted 30 lines
Quoted 27 lines
drivers/acpi/riscv/rhct.c +24
-0+if (!rhct) +return ; + } + +if (cbom_size) + *cbom_size =0 ; + +if (cboz_size) + *cboz_size =0 ; + +if (cbop_size) + *cbop_size =0 ; + + end = ACPI_ADD_PTR(struct acpi_rhct_node_header, rhct, rhct->header.length); + for (node = ACPI_ADD_PTR(struct acpi_rhct_node_header, rhct, rhct->node_offset); + node < end; + node = ACPI_ADD_PTR(struct acpi_rhct_node_header, node, node->length)) { +if (node->type ==ACPI_RHCT_NODE_TYPE_HART_INFO ) { + hart_info = ACPI_ADD_PTR(struct acpi_rhct_hart_info, node, size_hdr); +acpi_parse_hart_info_cmo_node (rhct, hart_info, cbom_size, + cboz_size, cbop_size); + } + } +}Signature · 1 line
2.39.2Other than the nits
Quoted 2 lines
Reviewed-by Andrew Jones <[email protected]>
Thanks a lot for the review!
Sunil
Re: [PATCH v3 -next 2/3] RISC-V: ACPI: RHCT: Add function to get CBO block sizes
Quoted 79 lines · Sunil V L
Quoted 69 lines · Andrew Jones
Quoted 68 lines · Sunil V L
Cache Block Operation (CBO) related block size in ACPI is provided by RHCT. Add support to read the CMO node in RHCT to get this information.
Signed-off-by Sunil V L <[email protected]>--- arch/riscv/include/asm/acpi.h | 6 +++ drivers/acpi/riscv/rhct.c | 93 +++++++++++++++++++++++++++++++++++ 2 files changed, 99 insertions(+)arch/riscv/include/asm/acpi.h +6
-0@@ -66,6 +66,8 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const char **isa);static inline int acpi_numa_get_nid (unsigned int cpu) {return NUMA_NO_NODE ; } +void acpi_get_cbo_block_size (struct acpi_table_header *table,u32 *cbom_size, +u32 *cboz_size,u32 *cbop_size);#else static inline void acpi_init_rintc_map (void ) { }static inline struct acpi_madt_rintc *acpi_cpu_get_madt_rintc (int cpu)@@ -79,6 +81,10 @@ static inline int acpi_get_riscv_isa(struct acpi_table_header *table,return -EINVAL ; } +static inline void acpi_get_cbo_block_size (struct acpi_table_header *table, +u32 *cbom_size,u32 *cboz_size, +u32 *cbop_size) { } +#endif /* CONFIG_ACPI */ #endif /*_ASM_ACPI_H*/ drivers/acpi/riscv/rhct.c +20
-0@@ -8,6 +8,7 @@#define pr_fmt (fmt) "ACPI: RHCT: " fmt#include <linux/acpi.h> +#include <linux/bits.h> static struct acpi_table_header *acpi_get_rhct (void ) {@@ -81,3 +82,95 @@ int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, constreturn -1 ; } + +static void acpi_parse_hart_info_cmo_node (struct acpi_table_rhct *rhct, +struct acpi_rhct_hart_info *hart_info, +u32 *cbom_size,u32 *cboz_size,u32 *cbop_size) +{ +u32 size_hartinfo =sizeof (struct acpi_rhct_hart_info ); +u32 size_hdr =sizeof (struct acpi_rhct_node_header ); +struct acpi_rhct_node_header *ref_node; +struct acpi_rhct_cmo_node *cmo_node; +u32 *hart_info_node_offset; + + hart_info_node_offset =ACPI_ADD_PTR (u32, hart_info, size_hartinfo); +for (int i =0 ; i < hart_info->num_offsets; i++) { + ref_node = ACPI_ADD_PTR(struct acpi_rhct_node_header, + rhct, hart_info_node_offset[i]); +if (ref_node->type ==ACPI_RHCT_NODE_TYPE_CMO ) { + cmo_node = ACPI_ADD_PTR(struct acpi_rhct_cmo_node, + ref_node, size_hdr); +if (cbom_size && cmo_node->cbom_size <=30 ) {Quoted 9 lines
drivers/acpi/riscv/rhct.c int acpi_get_riscv_isa(struct acpi_table_header *table, unsigned int cpu, const · line 107
Quoted 4 lines
drivers/acpi/riscv/rhct.c +4
-0+if (!*cbom_size) { + *cbom_size =BIT (cmo_node->cbom_size); + }else if (*cbom_size != +BIT (cmo_node->cbom_size)) {No need to break the if line, we can go to 100 chars. And then, since both the if and else if arms only have single statements, all the {} can be dropped too. Same comment for cboz and cbop.
Quoted 3 lines
Yeah, it is a side effect of working on repos with different coding standards. It is interesting that checkpatch didn't recommend to remove the braces. Let me fix it in next revision. Thanks!.
I think checkpatch is OK with single lines that get broken into multiple lines having braces around them. In fact, it may even be preferred. My suggestion is to not break the line though, so checkpatch would likely complain after turning these lines into true single lines.
Thanks,
drew
Sunil V L
maintainerPATCH v3 3/3 RISC-V: cacheflush: Initialize CBO variables on ACPI systems
In reply to Sunil V LInitialize the CBO variables on ACPI based systems using information in RHCT.
--- arch/riscv/mm/cacheflush.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-)
arch/riscv/mm/cacheflush.c +19 -6
@@ -3,7 +3,9 @@* Copyright (C) 2017 SiFive */ +#include <linux/acpi.h> #include <linux/of.h> +#include <asm/acpi.h> #include <asm/cacheflush.h> #ifdef CONFIG_SMP
@@ -124,13 +126,24 @@ void __init riscv_init_cbo_blocksizes(void)unsigned long cbom_hartid, cboz_hartid;u32 cbom_block_size =0 , cboz_block_size =0 ;struct device_node *node; +struct acpi_table_header *rhct; +acpi_status status; + +if (acpi_disabled) { +for_each_of_cpu_node (node) { +/* set block-size for cbom and/or cboz extension if available */ +cbo_get_block_size (node,"riscv,cbom-block-size" , + &cbom_block_size, &cbom_hartid); +cbo_get_block_size (node,"riscv,cboz-block-size" , + &cboz_block_size, &cboz_hartid); + } + }else { + status =acpi_get_table (ACPI_SIG_RHCT ,0 , &rhct); +if (ACPI_FAILURE (status)) +return ; -for_each_of_cpu_node (node) { -/* set block-size for cbom and/or cboz extension if available */ -cbo_get_block_size (node,"riscv,cbom-block-size" , - &cbom_block_size, &cbom_hartid); -cbo_get_block_size (node,"riscv,cboz-block-size" , - &cboz_block_size, &cboz_hartid); +acpi_get_cbo_block_size (rhct, &cbom_block_size, &cboz_block_size,NULL ); +acpi_put_table ((struct acpi_table_header *)rhct); }if (cbom_block_size)
Signature · 1 line
2.39.2
Re: [PATCH v3 -next 3/3] RISC-V: cacheflush: Initialize CBO variables on ACPI systems
Quoted 56 lines · Sunil V L
Initialize the CBO variables on ACPI based systems using information in RHCT.
Signed-off-by Sunil V L <[email protected]>--- arch/riscv/mm/cacheflush.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-)arch/riscv/mm/cacheflush.c +19
-6@@ -3,7 +3,9 @@* Copyright (C) 2017 SiFive */ +#include <linux/acpi.h> #include <linux/of.h> +#include <asm/acpi.h> #include <asm/cacheflush.h> #ifdef CONFIG_SMP @@ -124,13 +126,24 @@ void __init riscv_init_cbo_blocksizes(void)unsigned long cbom_hartid, cboz_hartid;u32 cbom_block_size =0 , cboz_block_size =0 ;struct device_node *node; +struct acpi_table_header *rhct; +acpi_status status; + +if (acpi_disabled) { +for_each_of_cpu_node (node) { +/* set block-size for cbom and/or cboz extension if available */ +cbo_get_block_size (node,"riscv,cbom-block-size" , + &cbom_block_size, &cbom_hartid); +cbo_get_block_size (node,"riscv,cboz-block-size" , + &cboz_block_size, &cboz_hartid); + } + }else { + status =acpi_get_table (ACPI_SIG_RHCT ,0 , &rhct); +if (ACPI_FAILURE (status)) +return ; -for_each_of_cpu_node (node) { -/* set block-size for cbom and/or cboz extension if available */ -cbo_get_block_size (node,"riscv,cbom-block-size" , - &cbom_block_size, &cbom_hartid); -cbo_get_block_size (node,"riscv,cboz-block-size" , - &cboz_block_size, &cboz_hartid); +acpi_get_cbo_block_size (rhct, &cbom_block_size, &cboz_block_size,NULL ); +acpi_put_table ((struct acpi_table_header *)rhct); }if (cbom_block_size)Signature · 1 line
2.39.2
Re: [PATCH v3 -next 3/3] RISC-V: cacheflush: Initialize CBO variables on ACPI systems
In reply to Sunil V LQuoted 7 lines · Sunil V L
Initialize the CBO variables on ACPI based systems using information in RHCT.
Signed-off-by Sunil V L <[email protected]>--- arch/riscv/mm/cacheflush.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-)