mm: use clear_user_(high)page() for arch with special user folio handling
Subsystems (12)
Zi Yan
reviewerPATCH v2 1/1 mm: use clear_user_(high)page() for arch with special user folio handling
For architectures setting cpu_dcache_is_aliasing() to true, which require flushing cache, and arc, which changes folio->flags after clearing a user folio, __GFP_ZERO using only clear_page() is not enough to zero user folios and clear_user_(high)page() must be used. Otherwise, user data will be corrupted.
Fix it by always clearing user folios with clear_user_(high)page() when cpu_dcache_is_aliasing() is true or architecture is arc. Rename alloc_zeroed() to alloc_need_zeroing() and invert the logic to clarify its intend.
--- include/linux/highmem.h | 8 +++++++- include/linux/mm.h | 17 +++++++++++++++++ mm/huge_memory.c | 9 +++++---- mm/internal.h | 6 ------ mm/memory.c | 10 +++++----- 5 files changed, 34 insertions(+), 16 deletions(-)
include/linux/highmem.h +7 -1
@@ -224,7 +224,13 @@ static inlinestruct folio *vma_alloc_zeroed_movable_folio (struct vm_area_struct *vma,unsigned long vaddr) { -return vma_alloc_folio (GFP_HIGHUSER_MOVABLE | __GFP_ZERO,0 , vma, vaddr); +struct folio *folio; + + folio =vma_alloc_folio (GFP_HIGHUSER_MOVABLE ,0 , vma, vaddr); +if (folio &&alloc_need_zeroing ()) +clear_user_highpage (&folio->page, vaddr); + +return folio; }#endif
include/linux/mm.h +17 -0
@@ -31,6 +31,7 @@#include <linux/kasan.h> #include <linux/memremap.h> #include <linux/slab.h> +#include <linux/cacheinfo.h> struct mempolicy ;struct anon_vma ;
@@ -4175,6 +4176,22 @@ static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla }#endif +/* +* alloc_need_zeroing checks if a user folio from page allocator needs to be +* zeroed or not. +*/ +static inline bool alloc_need_zeroing (void ) +{ +/* +* for user folios, arch with cache aliasing requires cache flush and +* arc changes folio->flags, so always return false to make caller use +* clear_user_page()/clear_user_highpage() +*/ +return (cpu_dcache_is_aliasing () ||IS_ENABLED (CONFIG_ARC )) || + !static_branch_maybe (CONFIG_INIT_ON_ALLOC_DEFAULT_ON , + &init_on_alloc); +} +int arch_get_shadow_stack_status (struct task_struct *t,unsigned long __user *status);int arch_set_shadow_stack_status (struct task_struct *t,unsigned long status);int arch_lock_shadow_stack_status (struct task_struct *t,unsigned long status);
mm/huge_memory.c +5 -4
@@ -1176,11 +1176,12 @@ static struct folio *vma_alloc_anon_folio_pmd(struct vm_area_struct *vma,folio_throttle_swaprate (folio, gfp);/* -* When a folio is not zeroed during allocation (__GFP_ZERO not used), -* folio_zero_user() is used to make sure that the page corresponding -* to the faulting address will be hot in the cache after zeroing. +* When a folio is not zeroed during allocation (__GFP_ZERO not used) +* or user folios require special handling, folio_zero_user() is used to +* make sure that the page corresponding to the faulting address will be +* hot in the cache after zeroing. */ -if (!alloc_zeroed ()) +if (alloc_need_zeroing ())folio_zero_user (folio, addr);/* * The memory barrier inside __folio_mark_uptodate makes sure that
mm/internal.h +0 -6
@@ -1285,12 +1285,6 @@ void touch_pud(struct vm_area_struct *vma, unsigned long addr,void touch_pmd (struct vm_area_struct *vma,unsigned long addr,pmd_t *pmd,bool write); -static inline bool alloc_zeroed (void ) -{ -return static_branch_maybe (CONFIG_INIT_ON_ALLOC_DEFAULT_ON , - &init_on_alloc); -} -/* * Parses a string with mem suffixes into its order. Useful to parse kernel * parameters.
mm/memory.c +5 -5
@@ -4733,12 +4733,12 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf)folio_throttle_swaprate (folio, gfp);/* * When a folio is not zeroed during allocation -* (__GFP_ZERO not used), folio_zero_user() is used -* to make sure that the page corresponding to the -* faulting address will be hot in the cache after -* zeroing. +* (__GFP_ZERO not used) or user folios require special +* handling, folio_zero_user() is used to make sure +* that the page corresponding to the faulting address +* will be hot in the cache after zeroing. */ -if (!alloc_zeroed ()) +if (alloc_need_zeroing ())folio_zero_user (folio, vmf->address);return folio; }
Signature · 1 line
2.45.2
Quoted 61 lines · Zi Yan
For architectures setting
cpu_dcache_is_aliasing()to true, which require flushing cache, and arc, which changesfolio->flagsafter clearing a user folio,__GFP_ZEROusing onlyclear_page()is not enough to zero user folios and clear_user_(high)page() must be used. Otherwise, user data will be corrupted.Fix it by always clearing user folios with clear_user_(high)page() when
cpu_dcache_is_aliasing()is true or architecture is arc. Renamealloc_zeroed()toalloc_need_zeroing()and invert the logic to clarify its intend.Fixes 5708d96da20b ("mm: avoid zeroing user movable page twice with init_on_alloc=1")Reported-by Geert Uytterhoeven <[email protected]>Closes https://lore.kernel.org/linux-mm/CAMuHMdV1hRp_NtR5YnJo=HsfgKQeH91J537Gh4gKk3PFZhSkbA@mail.gmail.com/ open hereTested-by Geert Uytterhoeven <[email protected]>Signed-off-by Zi Yan <[email protected]>--- include/linux/highmem.h | 8 +++++++- include/linux/mm.h | 17 +++++++++++++++++ mm/huge_memory.c | 9 +++++---- mm/internal.h | 6 ------ mm/memory.c | 10 +++++----- 5 files changed, 34 insertions(+), 16 deletions(-)include/linux/highmem.h +7
-1@@ -224,7 +224,13 @@ static inlinestruct folio *vma_alloc_zeroed_movable_folio (struct vm_area_struct *vma,unsigned long vaddr) { -return vma_alloc_folio (GFP_HIGHUSER_MOVABLE | __GFP_ZERO,0 , vma, vaddr); +struct folio *folio; + + folio =vma_alloc_folio (GFP_HIGHUSER_MOVABLE ,0 , vma, vaddr); +if (folio &&alloc_need_zeroing ()) +clear_user_highpage (&folio->page, vaddr); + +return folio; }#endif include/linux/mm.h +3
-0@@ -31,6 +31,7 @@#include <linux/kasan.h> #include <linux/memremap.h> #include <linux/slab.h> +#include <linux/cacheinfo.h> struct mempolicy ;struct anon_vma ;@@ -4175,6 +4176,22 @@ static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla }#endif +/* +* alloc_need_zeroing checks if a user folio from page allocator needs to be
include/linux/mm.h static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla · line 4190
Quoted 10 lines
include/linux/mm.h +10
-0+* zeroed or not. +*/ +static inline bool alloc_need_zeroing (void ) +{ +/* +* for user folios, arch with cache aliasing requires cache flush and +* arc changes folio->flags, so always return false to make caller use +* clear_user_page()/clear_user_highpage() +*/ + return (cpu_dcache_is_aliasing() || IS_ENABLED(CONFIG_ARC)) ||
Nack.
Can we please not go back to re-introducing arch-specific conditionals in generic mm code after the cleanup I did when introducing cpu_dcache_is_aliasing() in commit 8690bbcf3b70 ?
Based on commit eacd0e950dc2, AFAIU what you appear to need here is to introduce a:
cpu_icache_is_aliasing() -> note the "i" for instruction cache
It would typically be directly set to
#define cpu_icache_is_aliasing() cpu_dcache_is_aliasing()
except on architecture like ARC when the icache vs dcache is aliasing, but not dcache vs dcache.
So for ARC it would be defined as:
#define cpu_dcache_is_aliasing() false
#define cpu_icache_is_aliasing() true
And the Kconfig ARCH_HAS_CPU_CACHE_ALIASING=y would be set for ARC again.
I'm not entirely sure if we want to go for the wording "is_aliasing" or "is_incoherent" when talking about icache vs dcache, so I'm open to ideas here.
Thanks,
Mathieu
Quoted 67 lines
include/linux/mm.h +4
-0+ !static_branch_maybe (CONFIG_INIT_ON_ALLOC_DEFAULT_ON , + &init_on_alloc); +} +int arch_get_shadow_stack_status (struct task_struct *t,unsigned long __user *status);int arch_set_shadow_stack_status (struct task_struct *t,unsigned long status);int arch_lock_shadow_stack_status (struct task_struct *t,unsigned long status);mm/huge_memory.c +5
-4@@ -1176,11 +1176,12 @@ static struct folio *vma_alloc_anon_folio_pmd(struct vm_area_struct *vma,folio_throttle_swaprate (folio, gfp);/* -* When a folio is not zeroed during allocation (__GFP_ZERO not used), -* folio_zero_user() is used to make sure that the page corresponding -* to the faulting address will be hot in the cache after zeroing. +* When a folio is not zeroed during allocation (__GFP_ZERO not used) +* or user folios require special handling, folio_zero_user() is used to +* make sure that the page corresponding to the faulting address will be +* hot in the cache after zeroing. */ -if (!alloc_zeroed ()) +if (alloc_need_zeroing ())folio_zero_user (folio, addr);/* * The memory barrier inside __folio_mark_uptodate makes sure that mm/internal.h +0
-6@@ -1285,12 +1285,6 @@ void touch_pud(struct vm_area_struct *vma, unsigned long addr,void touch_pmd (struct vm_area_struct *vma,unsigned long addr,pmd_t *pmd,bool write); -static inline bool alloc_zeroed (void ) -{ -return static_branch_maybe (CONFIG_INIT_ON_ALLOC_DEFAULT_ON , - &init_on_alloc); -} -/* * Parses a string with mem suffixes into its order. Useful to parse kernel * parameters. mm/memory.c +5
-5@@ -4733,12 +4733,12 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf)folio_throttle_swaprate (folio, gfp);/* * When a folio is not zeroed during allocation -* (__GFP_ZERO not used), folio_zero_user() is used -* to make sure that the page corresponding to the -* faulting address will be hot in the cache after -* zeroing. +* (__GFP_ZERO not used) or user folios require special +* handling, folio_zero_user() is used to make sure +* that the page corresponding to the faulting address +* will be hot in the cache after zeroing. */ -if (!alloc_zeroed ()) +if (alloc_need_zeroing ())folio_zero_user (folio, vmf->address);return folio; }
Signature · 3 lines
Mathieu Desnoyers EfficiOS Inc. https://www.efficios.com
Zi Yan
reviewerQuoted 72 lines · Mathieu Desnoyers
Quoted 58 lines · Zi Yan
For architectures setting
cpu_dcache_is_aliasing()to true, which require flushing cache, and arc, which changesfolio->flagsafter clearing a user folio,__GFP_ZEROusing onlyclear_page()is not enough to zero user folios and clear_user_(high)page() must be used. Otherwise, user data will be corrupted.Fix it by always clearing user folios with clear_user_(high)page() when
cpu_dcache_is_aliasing()is true or architecture is arc. Renamealloc_zeroed()toalloc_need_zeroing()and invert the logic to clarify its intend.Fixes 5708d96da20b ("mm: avoid zeroing user movable page twice with init_on_alloc=1")Reported-by Geert Uytterhoeven <[email protected]>Closes https://lore.kernel.org/linux-mm/CAMuHMdV1hRp_NtR5YnJo=HsfgKQeH91J537Gh4gKk3PFZhSkbA@mail.gmail.com/ open hereTested-by Geert Uytterhoeven <[email protected]>Signed-off-by Zi Yan <[email protected]>--- include/linux/highmem.h | 8 +++++++- include/linux/mm.h | 17 +++++++++++++++++ mm/huge_memory.c | 9 +++++---- mm/internal.h | 6 ------ mm/memory.c | 10 +++++----- 5 files changed, 34 insertions(+), 16 deletions(-)include/linux/highmem.h +7
-1@@ -224,7 +224,13 @@ static inlinestruct folio *vma_alloc_zeroed_movable_folio (struct vm_area_struct *vma,unsigned long vaddr) { -return vma_alloc_folio (GFP_HIGHUSER_MOVABLE | __GFP_ZERO,0 , vma, vaddr); +struct folio *folio; + + folio =vma_alloc_folio (GFP_HIGHUSER_MOVABLE ,0 , vma, vaddr); +if (folio &&alloc_need_zeroing ()) +clear_user_highpage (&folio->page, vaddr); + +return folio; }#endif diff --git a/include/linux/mm.h b/include/linux/mm.hindex
c39c4945946c..ca8df5871213100644include/linux/mm.h +2
-0@@ -31,6 +31,7 @@#include <linux/kasan.h> #include <linux/memremap.h> #include <linux/slab.h> +#include <linux/cacheinfo.h> struct mempolicy ;struct anon_vma ;@@ -4175,6 +4176,22 @@ static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla }#endif +/* +* alloc_need_zeroing checks if a user folio from page allocator needs to be include/linux/mm.h static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla · line 4189
Quoted 10 lines
include/linux/mm.h +10
-0+* zeroed or not. +*/ +static inline bool alloc_need_zeroing (void ) +{ +/* +* for user folios, arch with cache aliasing requires cache flush and +* arc changes folio->flags, so always return false to make caller use +* clear_user_page()/clear_user_highpage() +*/ + return (cpu_dcache_is_aliasing() || IS_ENABLED(CONFIG_ARC)) ||Nack.
Quoted 3 lines
Can we please not go back to re-introducing arch-specific conditionals in generic mm code after the cleanup I did when introducing
cpu_dcache_is_aliasing()in commit8690bbcf3b70?
OK
Quoted 13 lines
Based on commit
eacd0e950dc2, AFAIU what you appear to need here is to introduce a:
cpu_icache_is_aliasing()-> note the "i" for instruction cacheIt would typically be directly set to
#define
cpu_icache_is_aliasing()cpu_dcache_is_aliasing()except on architecture like ARC when the icache vs dcache is aliasing, but not dcache vs dcache.
Quoted 8 lines
So for ARC it would be defined as:
#define
cpu_dcache_is_aliasing()false
#definecpu_icache_is_aliasing()trueAnd the
KconfigARCH_HAS_CPU_CACHE_ALIASING=y would be set for ARC again.
Sounds good to me.
Quoted 4 lines
I'm not entirely sure if we want to go for the wording "
is_aliasing" or "is_incoherent" when talking about icache vs dcache, so I'm open to ideas here.
Let me know if the code below looks good to you. I will use (cpu_icache_is_aliasing() || cpu_dcache_is_aliasing()) instead of
(cpu_dcache_is_aliasing() || IS_ENABLED(CONFIG_ARC)) in my next version.
arch/arc/Kconfig +1 -0
@@ -6,6 +6,7 @@ config ARC def_bool y select ARC_TIMERS + select ARCH_HAS_CPU_CACHE_ALIASING select ARCH_HAS_CACHE_LINE_SIZE select ARCH_HAS_DEBUG_VM_PGTABLE select ARCH_HAS_DMA_PREP_COHERENT
arch/arc/include/asm/cachetype.h new file +8 -0
@@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __ASM_ARC_CACHETYPE_H +#define __ASM_ARC_CACHETYPE_H + +#define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () true + +#endif
arch/arm/include/asm/cachetype.h +1 -0
@@ -21,6 +21,7 @@ extern unsigned int cacheid;#define icache_is_pipt () cacheid_is(CACHEID_PIPT)#define cpu_dcache_is_aliasing () (cache_is_vivt() || cache_is_vipt_aliasing()) +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()/* * __LINUX_ARM_ARCH__ is the minimum supported CPU architecture
arch/csky/include/asm/cachetype.h +1 -0
@@ -5,5 +5,6 @@#include <linux/types.h> #define cpu_dcache_is_aliasing () true +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/m68k/include/asm/cachetype.h +1 -0
@@ -5,5 +5,6 @@#include <linux/types.h> #define cpu_dcache_is_aliasing () true +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/mips/include/asm/cachetype.h +1 -0
@@ -5,5 +5,6 @@#include <asm/cpu-features.h> #define cpu_dcache_is_aliasing () cpu_has_dc_aliases +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/nios2/include/asm/cachetype.h +1 -0
@@ -6,5 +6,6 @@#include <asm/cache.h> #define cpu_dcache_is_aliasing () (NIOS2_DCACHE_SIZE > PAGE_SIZE) +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/parisc/include/asm/cachetype.h +1 -0
@@ -5,5 +5,6 @@#include <linux/types.h> #define cpu_dcache_is_aliasing () true +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/sh/include/asm/cachetype.h +1 -0
@@ -5,5 +5,6 @@#include <linux/types.h> #define cpu_dcache_is_aliasing () true +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
arch/xtensa/include/asm/cachetype.h +1 -0
@@ -6,5 +6,6 @@#include <asm/page.h> #define cpu_dcache_is_aliasing () (DCACHE_WAY_SIZE > PAGE_SIZE) +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#endif
include/linux/cacheinfo.h +1 -0
@@ -155,6 +155,7 @@ static inline int get_cpu_cacheinfo_id(int cpu, int level)#ifndef CONFIG_ARCH_HAS_CPU_CACHE_ALIASING #define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#else #include <asm/cachetype.h> #endif
Best Regards,
Yan, Zi
Zi Yan
reviewerQuoted 109 lines · Zi Yan
Quoted 72 lines · Mathieu Desnoyers
Quoted 58 lines · Zi Yan
For architectures setting
cpu_dcache_is_aliasing()to true, which require flushing cache, and arc, which changesfolio->flagsafter clearing a user folio,__GFP_ZEROusing onlyclear_page()is not enough to zero user folios and clear_user_(high)page() must be used. Otherwise, user data will be corrupted.Fix it by always clearing user folios with clear_user_(high)page() when
cpu_dcache_is_aliasing()is true or architecture is arc. Renamealloc_zeroed()toalloc_need_zeroing()and invert the logic to clarify its intend.Fixes 5708d96da20b ("mm: avoid zeroing user movable page twice with init_on_alloc=1")Reported-by Geert Uytterhoeven <[email protected]>Closes https://lore.kernel.org/linux-mm/CAMuHMdV1hRp_NtR5YnJo=HsfgKQeH91J537Gh4gKk3PFZhSkbA@mail.gmail.com/ open hereTested-by Geert Uytterhoeven <[email protected]>Signed-off-by Zi Yan <[email protected]>--- include/linux/highmem.h | 8 +++++++- include/linux/mm.h | 17 +++++++++++++++++ mm/huge_memory.c | 9 +++++---- mm/internal.h | 6 ------ mm/memory.c | 10 +++++----- 5 files changed, 34 insertions(+), 16 deletions(-)include/linux/highmem.h +7
-1@@ -224,7 +224,13 @@ static inlinestruct folio *vma_alloc_zeroed_movable_folio (struct vm_area_struct *vma,unsigned long vaddr) { -return vma_alloc_folio (GFP_HIGHUSER_MOVABLE | __GFP_ZERO,0 , vma, vaddr); +struct folio *folio; + + folio =vma_alloc_folio (GFP_HIGHUSER_MOVABLE ,0 , vma, vaddr); +if (folio &&alloc_need_zeroing ()) +clear_user_highpage (&folio->page, vaddr); + +return folio; }#endif diff --git a/include/linux/mm.h b/include/linux/mm.hindex
c39c4945946c..ca8df5871213100644include/linux/mm.h +2
-0@@ -31,6 +31,7 @@#include <linux/kasan.h> #include <linux/memremap.h> #include <linux/slab.h> +#include <linux/cacheinfo.h> struct mempolicy ;struct anon_vma ;@@ -4175,6 +4176,22 @@ static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla }#endif +/* +* alloc_need_zeroing checks if a user folio from page allocator needs to be include/linux/mm.h static inline int do_mseal(unsigned long start, size_t len_in, unsigned long fla · line 4189
Quoted 10 lines
include/linux/mm.h +10
-0+* zeroed or not. +*/ +static inline bool alloc_need_zeroing (void ) +{ +/* +* for user folios, arch with cache aliasing requires cache flush and +* arc changes folio->flags, so always return false to make caller use +* clear_user_page()/clear_user_highpage() +*/ + return (cpu_dcache_is_aliasing() || IS_ENABLED(CONFIG_ARC)) ||Nack.
Quoted 3 lines
Can we please not go back to re-introducing arch-specific conditionals in generic mm code after the cleanup I did when introducing
cpu_dcache_is_aliasing()in commit8690bbcf3b70?OK
Quoted 13 lines
Based on commit
eacd0e950dc2, AFAIU what you appear to need here is to introduce a:
cpu_icache_is_aliasing()-> note the "i" for instruction cacheIt would typically be directly set to
#define
cpu_icache_is_aliasing()cpu_dcache_is_aliasing()except on architecture like ARC when the icache vs dcache is aliasing, but not dcache vs dcache.
Quoted 8 lines
So for ARC it would be defined as:
#define
cpu_dcache_is_aliasing()false
#definecpu_icache_is_aliasing()trueAnd the
KconfigARCH_HAS_CPU_CACHE_ALIASING=y would be set for ARC again.Sounds good to me.
Quoted 4 lines
I'm not entirely sure if we want to go for the wording "
is_aliasing" or "is_incoherent" when talking about icache vs dcache, so I'm open to ideas here.
Quoted 5 lines
Let me know if the code below looks good to you. I will use (
cpu_icache_is_aliasing()||cpu_dcache_is_aliasing()) instead of
(cpu_dcache_is_aliasing()||IS_ENABLED(CONFIG_ARC)) in my next version.
Or this one, which set cpu_icache_is_aliasing()’s default value in the generic header:
arch/arc/Kconfig +1 -0
@@ -6,6 +6,7 @@ config ARC def_bool y select ARC_TIMERS + select ARCH_HAS_CPU_CACHE_ALIASING select ARCH_HAS_CACHE_LINE_SIZE select ARCH_HAS_DEBUG_VM_PGTABLE select ARCH_HAS_DMA_PREP_COHERENT
arch/arc/include/asm/cachetype.h new file +8 -0
@@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __ASM_ARC_CACHETYPE_H +#define __ASM_ARC_CACHETYPE_H + +#define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () true + +#endif
include/linux/cacheinfo.h +6 -0
@@ -155,8 +155,14 @@ static inline int get_cpu_cacheinfo_id(int cpu, int level)#ifndef CONFIG_ARCH_HAS_CPU_CACHE_ALIASING #define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#else #include <asm/cachetype.h> + +#ifndef cpu_icache_is_aliasing +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing() +#endif +#endif #endif /* _LINUX_CACHEINFO_H */
Best Regards,
Yan, Zi
Quoted 1 line · Zi Yan
On 7 Dec 2024, at 11:20, Zi Yan wrote:
[...]
Quoted 3 lines
Or this one, which set cpu_icache_is_aliasing()’s default value in the generic header:
Yes, this looks good.
Just be careful to keep the same tabs vs spaces style as the nearby code.
Thanks!
Mathieu
Quoted 50 lines
arch/arc/Kconfig +1
-0@@ -6,6 +6,7 @@ config ARC def_bool y select ARC_TIMERS + select ARCH_HAS_CPU_CACHE_ALIASING select ARCH_HAS_CACHE_LINE_SIZE select ARCH_HAS_DEBUG_VM_PGTABLE select ARCH_HAS_DMA_PREP_COHERENTarch/arc/include/asm/cachetype.h new file +8
-0@@ -0,0 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __ASM_ARC_CACHETYPE_H +#define __ASM_ARC_CACHETYPE_H + +#define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () true + +#endif include/linux/cacheinfo.h +6
-0@@ -155,8 +155,14 @@ static inline int get_cpu_cacheinfo_id(int cpu, int level)#ifndef CONFIG_ARCH_HAS_CPU_CACHE_ALIASING #define cpu_dcache_is_aliasing () false +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing()#else #include <asm/cachetype.h> + +#ifndef cpu_icache_is_aliasing +#define cpu_icache_is_aliasing () cpu_dcache_is_aliasing() +#endif +#endif #endif /* _LINUX_CACHEINFO_H */ Best Regards,
Yan, Zi
Signature · 3 lines
Mathieu Desnoyers EfficiOS Inc. https://www.efficios.com
Quoted 1 line · Zi Yan
On 7 Dec 2024, at 10:31, Mathieu Desnoyers wrote:
[...]
Quoted 10 lines
Quoted 4 lines
I'm not entirely sure if we want to go for the wording "
is_aliasing" or "is_incoherent" when talking about icache vs dcache, so I'm open to ideas here.Let me know if the code below looks good to you. I will use (
cpu_icache_is_aliasing()||cpu_dcache_is_aliasing()) instead of
(cpu_dcache_is_aliasing()||IS_ENABLED(CONFIG_ARC)) in my next version.
Yes, this is in line with what I have in mind.
And I prefer your second version with the generic cpu_icache_is_aliasing rather than defining it for each architecture.
Thanks,
Mathieu
Signature · 3 lines
Mathieu Desnoyers EfficiOS Inc. https://www.efficios.com