protection_map[] is an array based construct that translates given vm_flags
combination. This array contains page protection map, which is populated by
the platform via [__S000 .. __S111] and [__P000 .. __P111] exported macros.
Primary usage for protection_map[] is for vm_get_page_prot(), which is used
to determine page protection value for a given vm_flags. vm_get_page_prot()
implementation, could again call platform overrides arch_vm_get_page_prot()
and arch_filter_pgprot(). Some platforms override protection_map[] that was
originally built with __SXXX/__PXXX with different runtime values.
Currently there are multiple layers of abstraction i.e __SXXX/__PXXX macros
, protection_map[], arch_vm_get_page_prot() and arch_filter_pgprot() built
between the platform and generic MM, finally defining vm_get_page_prot().
Hence this series proposes to drop later two abstraction levels and instead
just move the responsibility of defining vm_get_page_prot() to the platform
(still utilizing generic protection_map[] array) itself making it clean and
simple.
This first introduces ARCH_HAS_VM_GET_PAGE_PROT which enables the platforms
to define custom vm_get_page_prot(). This starts converting platforms that
define the overrides arch_filter_pgprot() or arch_vm_get_page_prot() which
enables for those constructs to be dropped off completely.
The series has been inspired from an earlier discuss with Christoph Hellwig
https://lore.kernel.org/all/[email protected]/
This series applies on 5.18-rc1 after the following patch.
https://lore.kernel.org/all/[email protected]/
This series has been cross built for multiple platforms.
- Anshuman
Cc: Christoph Hellwig <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Changes in V4:
- ARCH_HAS_VM_GET_PAGE_PROT now excludes generic protection_map[]
- Changed platform's vm_get_page_prot() to use generic protection_map[]
- Dropped all platform changes not enabling either arch_vm_get_page_prot() or arch_filter_pgprot()
- Dropped all previous tags as code base has changed
Changes in V3:
https://lore.kernel.org/all/[email protected]/
- Dropped variable 'i' from sme_early_init() on x86 platform
- Moved CONFIG_COLDFIRE vm_get_page_prot() inside arch/m68k/mm/mcfmmu.c
- Moved CONFIG_SUN3 vm_get_page_prot() inside arch/m68k/mm/sun3mmu.c
- Dropped cachebits for vm_get_page_prot() inside arch/m68k/mm/motorola.c
- Dropped PAGE_XXX_C definitions from arch/m68k/include/asm/motorola_pgtable.h
- Used PAGE_XXX instead for vm_get_page_prot() inside arch/m68k/mm/motorola.c
- Dropped all references to protection_map[] in the tree
- Replaced s/extensa/xtensa/ on the patch title
- Moved left over comments from pgtable.h into init.c on nios2 platform
Changes in V2:
https://lore.kernel.org/all/[email protected]/
- Dropped the entire comment block in [PATCH 30/30] per Geert
- Replaced __P010 (although commented) with __PAGE_COPY on arm platform
- Replaced __P101 with PAGE_READONLY on um platform
Changes in V1:
https://lore.kernel.org/all/[email protected]/
- Add white spaces around the | operators
- Moved powerpc_vm_get_page_prot() near vm_get_page_prot() on powerpc
- Moved arm64_vm_get_page_prot() near vm_get_page_prot() on arm64
- Moved sparc_vm_get_page_prot() near vm_get_page_prot() on sparc
- Compacted vm_get_page_prot() switch cases on all platforms
- _PAGE_CACHE040 inclusion is dependent on CPU_IS_040_OR_060
- VM_SHARED case should return PAGE_NONE (not PAGE_COPY) on SH platform
- Reorganized VM_SHARED, VM_EXEC, VM_WRITE, VM_READ
- Dropped the last patch [RFC V1 31/31] which added macros for vm_flags combinations
https://lore.kernel.org/all/[email protected]/
Changes in RFC:
https://lore.kernel.org/all/[email protected]/
Anshuman Khandual (6):
mm/mmap: Add new config ARCH_HAS_VM_GET_PAGE_PROT
powerpc/mm: Enable ARCH_HAS_VM_GET_PAGE_PROT
arm64/mm: Enable ARCH_HAS_VM_GET_PAGE_PROT
sparc/mm: Enable ARCH_HAS_VM_GET_PAGE_PROT
mm/mmap: Drop arch_filter_pgprot()
mm/mmap: Drop arch_vm_get_page_pgprot()
Christoph Hellwig (1):
x86/mm: Enable ARCH_HAS_VM_GET_PAGE_PROT
arch/arm64/Kconfig | 1 +
arch/arm64/include/asm/mman.h | 24 ----------------------
arch/arm64/mm/mmap.c | 33 ++++++++++++++++++++++++++++++
arch/powerpc/Kconfig | 1 +
arch/powerpc/include/asm/mman.h | 12 -----------
arch/powerpc/mm/mmap.c | 26 ++++++++++++++++++++++++
arch/sparc/Kconfig | 1 +
arch/sparc/include/asm/mman.h | 6 ------
arch/sparc/mm/init_64.c | 13 ++++++++++++
arch/x86/Kconfig | 2 +-
arch/x86/include/asm/pgtable.h | 5 -----
arch/x86/include/uapi/asm/mman.h | 14 -------------
arch/x86/mm/Makefile | 2 +-
arch/x86/mm/pgprot.c | 35 ++++++++++++++++++++++++++++++++
include/linux/mman.h | 4 ----
mm/Kconfig | 2 +-
mm/mmap.c | 14 ++++---------
17 files changed, 117 insertions(+), 78 deletions(-)
create mode 100644 arch/x86/mm/pgprot.c
--
2.25.1
This defines and exports a platform specific custom vm_get_page_prot() via
subscribing ARCH_HAS_VM_GET_PAGE_PROT. It localizes arch_vm_get_page_prot()
as sparc_vm_get_page_prot() and moves near vm_get_page_prot().
Cc: "David S. Miller" <[email protected]>
Cc: Khalid Aziz <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Anshuman Khandual <[email protected]>
---
arch/sparc/Kconfig | 1 +
arch/sparc/include/asm/mman.h | 6 ------
arch/sparc/mm/init_64.c | 13 +++++++++++++
3 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 9200bc04701c..85b573643af6 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -84,6 +84,7 @@ config SPARC64
select PERF_USE_VMALLOC
select ARCH_HAVE_NMI_SAFE_CMPXCHG
select HAVE_C_RECORDMCOUNT
+ select ARCH_HAS_VM_GET_PAGE_PROT
select HAVE_ARCH_AUDITSYSCALL
select ARCH_SUPPORTS_ATOMIC_RMW
select ARCH_SUPPORTS_DEBUG_PAGEALLOC
diff --git a/arch/sparc/include/asm/mman.h b/arch/sparc/include/asm/mman.h
index 274217e7ed70..af9c10c83dc5 100644
--- a/arch/sparc/include/asm/mman.h
+++ b/arch/sparc/include/asm/mman.h
@@ -46,12 +46,6 @@ static inline unsigned long sparc_calc_vm_prot_bits(unsigned long prot)
}
}
-#define arch_vm_get_page_prot(vm_flags) sparc_vm_get_page_prot(vm_flags)
-static inline pgprot_t sparc_vm_get_page_prot(unsigned long vm_flags)
-{
- return (vm_flags & VM_SPARC_ADI) ? __pgprot(_PAGE_MCD_4V) : __pgprot(0);
-}
-
#define arch_validate_prot(prot, addr) sparc_validate_prot(prot, addr)
static inline int sparc_validate_prot(unsigned long prot, unsigned long addr)
{
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index 8b1911591581..dcb17763c1f2 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -3184,3 +3184,16 @@ void copy_highpage(struct page *to, struct page *from)
}
}
EXPORT_SYMBOL(copy_highpage);
+
+static pgprot_t sparc_vm_get_page_prot(unsigned long vm_flags)
+{
+ return (vm_flags & VM_SPARC_ADI) ? __pgprot(_PAGE_MCD_4V) : __pgprot(0);
+}
+
+pgprot_t vm_get_page_prot(unsigned long vm_flags)
+{
+ return __pgprot(pgprot_val(protection_map[vm_flags &
+ (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
+ pgprot_val(sparc_vm_get_page_prot(vm_flags)));
+}
+EXPORT_SYMBOL(vm_get_page_prot);
--
2.25.1
There are no platforms left which subscribe ARCH_HAS_FILTER_PGPROT. Hence
drop generic arch_filter_pgprot() and also config ARCH_HAS_FILTER_PGPROT.
Cc: Andrew Morton <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Anshuman Khandual <[email protected]>
---
mm/Kconfig | 3 ---
mm/mmap.c | 9 +--------
2 files changed, 1 insertion(+), 11 deletions(-)
diff --git a/mm/Kconfig b/mm/Kconfig
index b1f7624276f8..3f7b6d7b69df 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -762,9 +762,6 @@ config ARCH_HAS_CURRENT_STACK_POINTER
register alias named "current_stack_pointer", this config can be
selected.
-config ARCH_HAS_FILTER_PGPROT
- bool
-
config ARCH_HAS_VM_GET_PAGE_PROT
bool
diff --git a/mm/mmap.c b/mm/mmap.c
index 87cb2eaf7e1a..edf2a5e38f4d 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -107,20 +107,13 @@ pgprot_t protection_map[16] __ro_after_init = {
};
#ifndef CONFIG_ARCH_HAS_VM_GET_PAGE_PROT
-#ifndef CONFIG_ARCH_HAS_FILTER_PGPROT
-static inline pgprot_t arch_filter_pgprot(pgprot_t prot)
-{
- return prot;
-}
-#endif
-
pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
pgprot_t ret = __pgprot(pgprot_val(protection_map[vm_flags &
(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
pgprot_val(arch_vm_get_page_prot(vm_flags)));
- return arch_filter_pgprot(ret);
+ return ret;
}
EXPORT_SYMBOL(vm_get_page_prot);
#endif /* CONFIG_ARCH_HAS_VM_GET_PAGE_PROT */
--
2.25.1
From: Christoph Hellwig <[email protected]>
This defines and exports a platform specific custom vm_get_page_prot() via
subscribing ARCH_HAS_VM_GET_PAGE_PROT. This also unsubscribes from config
ARCH_HAS_FILTER_PGPROT, after dropping off arch_filter_pgprot() and
arch_vm_get_page_prot().
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: [email protected]
Signed-off-by: Christoph Hellwig <[email protected]>
Signed-off-by: Anshuman Khandual <[email protected]>
---
arch/x86/Kconfig | 2 +-
arch/x86/include/asm/pgtable.h | 5 -----
arch/x86/include/uapi/asm/mman.h | 14 -------------
arch/x86/mm/Makefile | 2 +-
arch/x86/mm/pgprot.c | 35 ++++++++++++++++++++++++++++++++
5 files changed, 37 insertions(+), 21 deletions(-)
create mode 100644 arch/x86/mm/pgprot.c
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index b0142e01002e..c355c420150e 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -76,7 +76,6 @@ config X86
select ARCH_HAS_EARLY_DEBUG if KGDB
select ARCH_HAS_ELF_RANDOMIZE
select ARCH_HAS_FAST_MULTIPLIER
- select ARCH_HAS_FILTER_PGPROT
select ARCH_HAS_FORTIFY_SOURCE
select ARCH_HAS_GCOV_PROFILE_ALL
select ARCH_HAS_KCOV if X86_64
@@ -95,6 +94,7 @@ config X86
select ARCH_HAS_SYNC_CORE_BEFORE_USERMODE
select ARCH_HAS_SYSCALL_WRAPPER
select ARCH_HAS_UBSAN_SANITIZE_ALL
+ select ARCH_HAS_VM_GET_PAGE_PROT
select ARCH_HAS_DEBUG_WX
select ARCH_HAS_ZONE_DMA_SET if EXPERT
select ARCH_HAVE_NMI_SAFE_CMPXCHG
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index 62ab07e24aef..3563f4645fa1 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -649,11 +649,6 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
#define canon_pgprot(p) __pgprot(massage_pgprot(p))
-static inline pgprot_t arch_filter_pgprot(pgprot_t prot)
-{
- return canon_pgprot(prot);
-}
-
static inline int is_new_memtype_allowed(u64 paddr, unsigned long size,
enum page_cache_mode pcm,
enum page_cache_mode new_pcm)
diff --git a/arch/x86/include/uapi/asm/mman.h b/arch/x86/include/uapi/asm/mman.h
index d4a8d0424bfb..775dbd3aff73 100644
--- a/arch/x86/include/uapi/asm/mman.h
+++ b/arch/x86/include/uapi/asm/mman.h
@@ -5,20 +5,6 @@
#define MAP_32BIT 0x40 /* only give out 32bit addresses */
#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
-/*
- * Take the 4 protection key bits out of the vma->vm_flags
- * value and turn them in to the bits that we can put in
- * to a pte.
- *
- * Only override these if Protection Keys are available
- * (which is only on 64-bit).
- */
-#define arch_vm_get_page_prot(vm_flags) __pgprot( \
- ((vm_flags) & VM_PKEY_BIT0 ? _PAGE_PKEY_BIT0 : 0) | \
- ((vm_flags) & VM_PKEY_BIT1 ? _PAGE_PKEY_BIT1 : 0) | \
- ((vm_flags) & VM_PKEY_BIT2 ? _PAGE_PKEY_BIT2 : 0) | \
- ((vm_flags) & VM_PKEY_BIT3 ? _PAGE_PKEY_BIT3 : 0))
-
#define arch_calc_vm_prot_bits(prot, key) ( \
((key) & 0x1 ? VM_PKEY_BIT0 : 0) | \
((key) & 0x2 ? VM_PKEY_BIT1 : 0) | \
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index fe3d3061fc11..fb6b41a48ae5 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -20,7 +20,7 @@ CFLAGS_REMOVE_mem_encrypt_identity.o = -pg
endif
obj-y := init.o init_$(BITS).o fault.o ioremap.o extable.o mmap.o \
- pgtable.o physaddr.o setup_nx.o tlb.o cpu_entry_area.o maccess.o
+ pgtable.o physaddr.o setup_nx.o tlb.o cpu_entry_area.o maccess.o pgprot.o
obj-y += pat/
diff --git a/arch/x86/mm/pgprot.c b/arch/x86/mm/pgprot.c
new file mode 100644
index 000000000000..763742782286
--- /dev/null
+++ b/arch/x86/mm/pgprot.c
@@ -0,0 +1,35 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/export.h>
+#include <linux/mm.h>
+#include <asm/pgtable.h>
+
+pgprot_t vm_get_page_prot(unsigned long vm_flags)
+{
+ unsigned long val = pgprot_val(protection_map[vm_flags &
+ (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]);
+
+#ifdef CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS
+ /*
+ * Take the 4 protection key bits out of the vma->vm_flags value and
+ * turn them in to the bits that we can put in to a pte.
+ *
+ * Only override these if Protection Keys are available (which is only
+ * on 64-bit).
+ */
+ if (vm_flags & VM_PKEY_BIT0)
+ val |= _PAGE_PKEY_BIT0;
+ if (vm_flags & VM_PKEY_BIT1)
+ val |= _PAGE_PKEY_BIT1;
+ if (vm_flags & VM_PKEY_BIT2)
+ val |= _PAGE_PKEY_BIT2;
+ if (vm_flags & VM_PKEY_BIT3)
+ val |= _PAGE_PKEY_BIT3;
+#endif
+
+ val = __sme_set(val);
+ if (val & _PAGE_PRESENT)
+ val &= __supported_pte_mask;
+ return __pgprot(val);
+}
+EXPORT_SYMBOL(vm_get_page_prot);
--
2.25.1
There are no platforms left which use arch_vm_get_page_prot(). Just drop
generic arch_vm_get_page_prot().
Cc: Andrew Morton <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Anshuman Khandual <[email protected]>
---
include/linux/mman.h | 4 ----
mm/mmap.c | 3 +--
2 files changed, 1 insertion(+), 6 deletions(-)
diff --git a/include/linux/mman.h b/include/linux/mman.h
index b66e91b8176c..58b3abd457a3 100644
--- a/include/linux/mman.h
+++ b/include/linux/mman.h
@@ -93,10 +93,6 @@ static inline void vm_unacct_memory(long pages)
#define arch_calc_vm_flag_bits(flags) 0
#endif
-#ifndef arch_vm_get_page_prot
-#define arch_vm_get_page_prot(vm_flags) __pgprot(0)
-#endif
-
#ifndef arch_validate_prot
/*
* This is called from mprotect(). PROT_GROWSDOWN and PROT_GROWSUP have
diff --git a/mm/mmap.c b/mm/mmap.c
index edf2a5e38f4d..db7f33154206 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -110,8 +110,7 @@ pgprot_t protection_map[16] __ro_after_init = {
pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
pgprot_t ret = __pgprot(pgprot_val(protection_map[vm_flags &
- (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
- pgprot_val(arch_vm_get_page_prot(vm_flags)));
+ (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]));
return ret;
}
--
2.25.1
This defines and exports a platform specific custom vm_get_page_prot() via
subscribing ARCH_HAS_VM_GET_PAGE_PROT. It localizes arch_vm_get_page_prot()
and moves it near vm_get_page_prot().
Cc: Catalin Marinas <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Anshuman Khandual <[email protected]>
---
arch/arm64/Kconfig | 1 +
arch/arm64/include/asm/mman.h | 24 ------------------------
arch/arm64/mm/mmap.c | 33 +++++++++++++++++++++++++++++++++
3 files changed, 34 insertions(+), 24 deletions(-)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 57c4c995965f..dd0b15162bb3 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -45,6 +45,7 @@ config ARM64
select ARCH_HAS_SYSCALL_WRAPPER
select ARCH_HAS_TEARDOWN_DMA_OPS if IOMMU_SUPPORT
select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
+ select ARCH_HAS_VM_GET_PAGE_PROT
select ARCH_HAS_ZONE_DMA_SET if EXPERT
select ARCH_HAVE_ELF_PROT
select ARCH_HAVE_NMI_SAFE_CMPXCHG
diff --git a/arch/arm64/include/asm/mman.h b/arch/arm64/include/asm/mman.h
index e3e28f7daf62..5966ee4a6154 100644
--- a/arch/arm64/include/asm/mman.h
+++ b/arch/arm64/include/asm/mman.h
@@ -35,30 +35,6 @@ static inline unsigned long arch_calc_vm_flag_bits(unsigned long flags)
}
#define arch_calc_vm_flag_bits(flags) arch_calc_vm_flag_bits(flags)
-static inline pgprot_t arch_vm_get_page_prot(unsigned long vm_flags)
-{
- pteval_t prot = 0;
-
- if (vm_flags & VM_ARM64_BTI)
- prot |= PTE_GP;
-
- /*
- * There are two conditions required for returning a Normal Tagged
- * memory type: (1) the user requested it via PROT_MTE passed to
- * mmap() or mprotect() and (2) the corresponding vma supports MTE. We
- * register (1) as VM_MTE in the vma->vm_flags and (2) as
- * VM_MTE_ALLOWED. Note that the latter can only be set during the
- * mmap() call since mprotect() does not accept MAP_* flags.
- * Checking for VM_MTE only is sufficient since arch_validate_flags()
- * does not permit (VM_MTE & !VM_MTE_ALLOWED).
- */
- if (vm_flags & VM_MTE)
- prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
-
- return __pgprot(prot);
-}
-#define arch_vm_get_page_prot(vm_flags) arch_vm_get_page_prot(vm_flags)
-
static inline bool arch_validate_prot(unsigned long prot,
unsigned long addr __always_unused)
{
diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
index 77ada00280d9..307534fcec00 100644
--- a/arch/arm64/mm/mmap.c
+++ b/arch/arm64/mm/mmap.c
@@ -55,3 +55,36 @@ static int __init adjust_protection_map(void)
return 0;
}
arch_initcall(adjust_protection_map);
+
+static pgprot_t arm64_arch_vm_get_page_prot(unsigned long vm_flags)
+{
+ pteval_t prot = 0;
+
+ if (vm_flags & VM_ARM64_BTI)
+ prot |= PTE_GP;
+
+ /*
+ * There are two conditions required for returning a Normal Tagged
+ * memory type: (1) the user requested it via PROT_MTE passed to
+ * mmap() or mprotect() and (2) the corresponding vma supports MTE. We
+ * register (1) as VM_MTE in the vma->vm_flags and (2) as
+ * VM_MTE_ALLOWED. Note that the latter can only be set during the
+ * mmap() call since mprotect() does not accept MAP_* flags.
+ * Checking for VM_MTE only is sufficient since arch_validate_flags()
+ * does not permit (VM_MTE & !VM_MTE_ALLOWED).
+ */
+ if (vm_flags & VM_MTE)
+ prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
+
+ return __pgprot(prot);
+}
+
+pgprot_t vm_get_page_prot(unsigned long vm_flags)
+{
+ pgprot_t ret = __pgprot(pgprot_val(protection_map[vm_flags &
+ (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
+ pgprot_val(arm64_arch_vm_get_page_prot(vm_flags)));
+
+ return ret;
+}
+EXPORT_SYMBOL(vm_get_page_prot);
--
2.25.1
Add a new config ARCH_HAS_VM_GET_PAGE_PROT, which when subscribed enables a
given platform to define its own vm_get_page_prot() but still utilizing the
generic protection_map[] array.
Cc: Andrew Morton <[email protected]>
Cc: [email protected]
Cc: [email protected]
Suggested-by: Christoph Hellwig <[email protected]>
Signed-off-by: Anshuman Khandual <[email protected]>
---
mm/Kconfig | 3 +++
mm/mmap.c | 2 ++
2 files changed, 5 insertions(+)
diff --git a/mm/Kconfig b/mm/Kconfig
index 034d87953600..b1f7624276f8 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -765,6 +765,9 @@ config ARCH_HAS_CURRENT_STACK_POINTER
config ARCH_HAS_FILTER_PGPROT
bool
+config ARCH_HAS_VM_GET_PAGE_PROT
+ bool
+
config ARCH_HAS_PTE_DEVMAP
bool
diff --git a/mm/mmap.c b/mm/mmap.c
index 3aa839f81e63..87cb2eaf7e1a 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -106,6 +106,7 @@ pgprot_t protection_map[16] __ro_after_init = {
__S000, __S001, __S010, __S011, __S100, __S101, __S110, __S111
};
+#ifndef CONFIG_ARCH_HAS_VM_GET_PAGE_PROT
#ifndef CONFIG_ARCH_HAS_FILTER_PGPROT
static inline pgprot_t arch_filter_pgprot(pgprot_t prot)
{
@@ -122,6 +123,7 @@ pgprot_t vm_get_page_prot(unsigned long vm_flags)
return arch_filter_pgprot(ret);
}
EXPORT_SYMBOL(vm_get_page_prot);
+#endif /* CONFIG_ARCH_HAS_VM_GET_PAGE_PROT */
static pgprot_t vm_pgprot_modify(pgprot_t oldprot, unsigned long vm_flags)
{
--
2.25.1
On 4/7/22 04:32, Anshuman Khandual wrote:
> This defines and exports a platform specific custom vm_get_page_prot() via
> subscribing ARCH_HAS_VM_GET_PAGE_PROT. It localizes arch_vm_get_page_prot()
> as sparc_vm_get_page_prot() and moves near vm_get_page_prot().
>
> Cc: "David S. Miller" <[email protected]>
> Cc: Khalid Aziz <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Signed-off-by: Anshuman Khandual <[email protected]>
> ---
> arch/sparc/Kconfig | 1 +
> arch/sparc/include/asm/mman.h | 6 ------
> arch/sparc/mm/init_64.c | 13 +++++++++++++
> 3 files changed, 14 insertions(+), 6 deletions(-)
>
Reviewed-by: Khalid Aziz <[email protected]>
> diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
> index 9200bc04701c..85b573643af6 100644
> --- a/arch/sparc/Kconfig
> +++ b/arch/sparc/Kconfig
> @@ -84,6 +84,7 @@ config SPARC64
> select PERF_USE_VMALLOC
> select ARCH_HAVE_NMI_SAFE_CMPXCHG
> select HAVE_C_RECORDMCOUNT
> + select ARCH_HAS_VM_GET_PAGE_PROT
> select HAVE_ARCH_AUDITSYSCALL
> select ARCH_SUPPORTS_ATOMIC_RMW
> select ARCH_SUPPORTS_DEBUG_PAGEALLOC
> diff --git a/arch/sparc/include/asm/mman.h b/arch/sparc/include/asm/mman.h
> index 274217e7ed70..af9c10c83dc5 100644
> --- a/arch/sparc/include/asm/mman.h
> +++ b/arch/sparc/include/asm/mman.h
> @@ -46,12 +46,6 @@ static inline unsigned long sparc_calc_vm_prot_bits(unsigned long prot)
> }
> }
>
> -#define arch_vm_get_page_prot(vm_flags) sparc_vm_get_page_prot(vm_flags)
> -static inline pgprot_t sparc_vm_get_page_prot(unsigned long vm_flags)
> -{
> - return (vm_flags & VM_SPARC_ADI) ? __pgprot(_PAGE_MCD_4V) : __pgprot(0);
> -}
> -
> #define arch_validate_prot(prot, addr) sparc_validate_prot(prot, addr)
> static inline int sparc_validate_prot(unsigned long prot, unsigned long addr)
> {
> diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
> index 8b1911591581..dcb17763c1f2 100644
> --- a/arch/sparc/mm/init_64.c
> +++ b/arch/sparc/mm/init_64.c
> @@ -3184,3 +3184,16 @@ void copy_highpage(struct page *to, struct page *from)
> }
> }
> EXPORT_SYMBOL(copy_highpage);
> +
> +static pgprot_t sparc_vm_get_page_prot(unsigned long vm_flags)
> +{
> + return (vm_flags & VM_SPARC_ADI) ? __pgprot(_PAGE_MCD_4V) : __pgprot(0);
> +}
> +
> +pgprot_t vm_get_page_prot(unsigned long vm_flags)
> +{
> + return __pgprot(pgprot_val(protection_map[vm_flags &
> + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
> + pgprot_val(sparc_vm_get_page_prot(vm_flags)));
> +}
> +EXPORT_SYMBOL(vm_get_page_prot);
On Thu, 7 Apr 2022 16:02:44 +0530 Anshuman Khandual <[email protected]> wrote:
> protection_map[] is an array based construct that translates given vm_flags
> combination. This array contains page protection map, which is populated by
> the platform via [__S000 .. __S111] and [__P000 .. __P111] exported macros.
> Primary usage for protection_map[] is for vm_get_page_prot(), which is used
> to determine page protection value for a given vm_flags. vm_get_page_prot()
> implementation, could again call platform overrides arch_vm_get_page_prot()
> and arch_filter_pgprot(). Some platforms override protection_map[] that was
> originally built with __SXXX/__PXXX with different runtime values.
>
> Currently there are multiple layers of abstraction i.e __SXXX/__PXXX macros
> , protection_map[], arch_vm_get_page_prot() and arch_filter_pgprot() built
> between the platform and generic MM, finally defining vm_get_page_prot().
>
> Hence this series proposes to drop later two abstraction levels and instead
> just move the responsibility of defining vm_get_page_prot() to the platform
> (still utilizing generic protection_map[] array) itself making it clean and
> simple.
>
> This first introduces ARCH_HAS_VM_GET_PAGE_PROT which enables the platforms
> to define custom vm_get_page_prot(). This starts converting platforms that
> define the overrides arch_filter_pgprot() or arch_vm_get_page_prot() which
> enables for those constructs to be dropped off completely.
>
> The series has been inspired from an earlier discuss with Christoph Hellwig
>
> https://lore.kernel.org/all/[email protected]/
>
> This series applies on 5.18-rc1 after the following patch.
>
> https://lore.kernel.org/all/[email protected]/
Confusing. That patch is already in 5.18-rc1.
But the version which was merged (24e988c7fd1ee701e) lacked the change
to arch/arm64/Kconfig. I seem to recall that this patch went through a
few issues and perhaps the arm64 change was dropped. Can you please
check?
(It would be easier for me to track all this down if the original patch
had had cc:linux-mm. Please cc linux-mm!)
On 4/8/22 04:50, Andrew Morton wrote:
> On Thu, 7 Apr 2022 16:02:44 +0530 Anshuman Khandual <[email protected]> wrote:
>
>> protection_map[] is an array based construct that translates given vm_flags
>> combination. This array contains page protection map, which is populated by
>> the platform via [__S000 .. __S111] and [__P000 .. __P111] exported macros.
>> Primary usage for protection_map[] is for vm_get_page_prot(), which is used
>> to determine page protection value for a given vm_flags. vm_get_page_prot()
>> implementation, could again call platform overrides arch_vm_get_page_prot()
>> and arch_filter_pgprot(). Some platforms override protection_map[] that was
>> originally built with __SXXX/__PXXX with different runtime values.
>>
>> Currently there are multiple layers of abstraction i.e __SXXX/__PXXX macros
>> , protection_map[], arch_vm_get_page_prot() and arch_filter_pgprot() built
>> between the platform and generic MM, finally defining vm_get_page_prot().
>>
>> Hence this series proposes to drop later two abstraction levels and instead
>> just move the responsibility of defining vm_get_page_prot() to the platform
>> (still utilizing generic protection_map[] array) itself making it clean and
>> simple.
>>
>> This first introduces ARCH_HAS_VM_GET_PAGE_PROT which enables the platforms
>> to define custom vm_get_page_prot(). This starts converting platforms that
>> define the overrides arch_filter_pgprot() or arch_vm_get_page_prot() which
>> enables for those constructs to be dropped off completely.
>>
>> The series has been inspired from an earlier discuss with Christoph Hellwig
>>
>> https://lore.kernel.org/all/[email protected]/
>>
>> This series applies on 5.18-rc1 after the following patch.
>>
>> https://lore.kernel.org/all/[email protected]/
>
> Confusing. That patch is already in 5.18-rc1.
Ahh, my bad, forgot to delete these lines here in the cover letter.
This series just applies cleanly on 5.18-rc1 without dependency.
> But the version which was merged (24e988c7fd1ee701e) lacked the change
> to arch/arm64/Kconfig. I seem to recall that this patch went through a
> few issues and perhaps the arm64 change was dropped. Can you please
> check?
ARCH_HAS_FILTER_PGPROT on arm64 got dropped off via another commit i.e
6e2edd6371a4 ("arm64: Ensure execute-only permissions are not allowed
without EPAN").
>
> (It would be easier for me to track all this down if the original patch
> had had cc:linux-mm. Please cc linux-mm!
Sure, will do. Please do let me know if there is anything else that needs
to be taken care.
On Thu, Apr 07, 2022 at 04:02:50PM +0530, Anshuman Khandual wrote:
> There are no platforms left which subscribe ARCH_HAS_FILTER_PGPROT. Hence
> drop generic arch_filter_pgprot() and also config ARCH_HAS_FILTER_PGPROT.
>
> Cc: Andrew Morton <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Signed-off-by: Anshuman Khandual <[email protected]>
Reviewed-by: Catalin Marinas <[email protected]>
On Thu, Apr 07, 2022 at 04:02:45PM +0530, Anshuman Khandual wrote:
> Add a new config ARCH_HAS_VM_GET_PAGE_PROT, which when subscribed enables a
> given platform to define its own vm_get_page_prot() but still utilizing the
> generic protection_map[] array.
>
> Cc: Andrew Morton <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Suggested-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Anshuman Khandual <[email protected]>
Reviewed-by: Catalin Marinas <[email protected]>
On Thu, Apr 07, 2022 at 04:02:47PM +0530, Anshuman Khandual wrote:
> diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
> index 77ada00280d9..307534fcec00 100644
> --- a/arch/arm64/mm/mmap.c
> +++ b/arch/arm64/mm/mmap.c
> @@ -55,3 +55,36 @@ static int __init adjust_protection_map(void)
> return 0;
> }
> arch_initcall(adjust_protection_map);
> +
> +static pgprot_t arm64_arch_vm_get_page_prot(unsigned long vm_flags)
> +{
> + pteval_t prot = 0;
> +
> + if (vm_flags & VM_ARM64_BTI)
> + prot |= PTE_GP;
> +
> + /*
> + * There are two conditions required for returning a Normal Tagged
> + * memory type: (1) the user requested it via PROT_MTE passed to
> + * mmap() or mprotect() and (2) the corresponding vma supports MTE. We
> + * register (1) as VM_MTE in the vma->vm_flags and (2) as
> + * VM_MTE_ALLOWED. Note that the latter can only be set during the
> + * mmap() call since mprotect() does not accept MAP_* flags.
> + * Checking for VM_MTE only is sufficient since arch_validate_flags()
> + * does not permit (VM_MTE & !VM_MTE_ALLOWED).
> + */
> + if (vm_flags & VM_MTE)
> + prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
> +
> + return __pgprot(prot);
> +}
> +
> +pgprot_t vm_get_page_prot(unsigned long vm_flags)
> +{
> + pgprot_t ret = __pgprot(pgprot_val(protection_map[vm_flags &
> + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
> + pgprot_val(arm64_arch_vm_get_page_prot(vm_flags)));
> +
> + return ret;
> +}
> +EXPORT_SYMBOL(vm_get_page_prot);
Could you write all this in a single function? I think I mentioned it in
a previous series (untested):
pgprot_t vm_get_page_prot(unsigned long vm_flags)
{
pteval_t prot = pgprot_val(protection_map[vm_flags &
(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]);
if (vm_flags & VM_ARM64_BTI)
prot |= PTE_GP;
/*
* There are two conditions required for returning a Normal Tagged
* memory type: (1) the user requested it via PROT_MTE passed to
* mmap() or mprotect() and (2) the corresponding vma supports MTE. We
* register (1) as VM_MTE in the vma->vm_flags and (2) as
* VM_MTE_ALLOWED. Note that the latter can only be set during the
* mmap() call since mprotect() does not accept MAP_* flags.
* Checking for VM_MTE only is sufficient since arch_validate_flags()
* does not permit (VM_MTE & !VM_MTE_ALLOWED).
*/
if (vm_flags & VM_MTE)
prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
return __pgprot(prot);
}
EXPORT_SYMBOL(vm_get_page_prot);
With that:
Reviewed-by: Catalin Marinas <[email protected]>
On Thu, Apr 07, 2022 at 04:02:51PM +0530, Anshuman Khandual wrote:
> There are no platforms left which use arch_vm_get_page_prot(). Just drop
> generic arch_vm_get_page_prot().
>
> Cc: Andrew Morton <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Signed-off-by: Anshuman Khandual <[email protected]>
Reviewed-by: Catalin Marinas <[email protected]>
On 4/8/22 15:58, Catalin Marinas wrote:
> On Thu, Apr 07, 2022 at 04:02:47PM +0530, Anshuman Khandual wrote:
>> diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
>> index 77ada00280d9..307534fcec00 100644
>> --- a/arch/arm64/mm/mmap.c
>> +++ b/arch/arm64/mm/mmap.c
>> @@ -55,3 +55,36 @@ static int __init adjust_protection_map(void)
>> return 0;
>> }
>> arch_initcall(adjust_protection_map);
>> +
>> +static pgprot_t arm64_arch_vm_get_page_prot(unsigned long vm_flags)
>> +{
>> + pteval_t prot = 0;
>> +
>> + if (vm_flags & VM_ARM64_BTI)
>> + prot |= PTE_GP;
>> +
>> + /*
>> + * There are two conditions required for returning a Normal Tagged
>> + * memory type: (1) the user requested it via PROT_MTE passed to
>> + * mmap() or mprotect() and (2) the corresponding vma supports MTE. We
>> + * register (1) as VM_MTE in the vma->vm_flags and (2) as
>> + * VM_MTE_ALLOWED. Note that the latter can only be set during the
>> + * mmap() call since mprotect() does not accept MAP_* flags.
>> + * Checking for VM_MTE only is sufficient since arch_validate_flags()
>> + * does not permit (VM_MTE & !VM_MTE_ALLOWED).
>> + */
>> + if (vm_flags & VM_MTE)
>> + prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
>> +
>> + return __pgprot(prot);
>> +}
>> +
>> +pgprot_t vm_get_page_prot(unsigned long vm_flags)
>> +{
>> + pgprot_t ret = __pgprot(pgprot_val(protection_map[vm_flags &
>> + (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]) |
>> + pgprot_val(arm64_arch_vm_get_page_prot(vm_flags)));
>> +
>> + return ret;
>> +}
>> +EXPORT_SYMBOL(vm_get_page_prot);
>
> Could you write all this in a single function? I think I mentioned it in
> a previous series (untested):
Right, missed that.
>
> pgprot_t vm_get_page_prot(unsigned long vm_flags)
> {
> pteval_t prot = pgprot_val(protection_map[vm_flags &
> (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]);
>
> if (vm_flags & VM_ARM64_BTI)
> prot |= PTE_GP;
>
> /*
> * There are two conditions required for returning a Normal Tagged
> * memory type: (1) the user requested it via PROT_MTE passed to
> * mmap() or mprotect() and (2) the corresponding vma supports MTE. We
> * register (1) as VM_MTE in the vma->vm_flags and (2) as
> * VM_MTE_ALLOWED. Note that the latter can only be set during the
> * mmap() call since mprotect() does not accept MAP_* flags.
> * Checking for VM_MTE only is sufficient since arch_validate_flags()
> * does not permit (VM_MTE & !VM_MTE_ALLOWED).
> */
> if (vm_flags & VM_MTE)
> prot |= PTE_ATTRINDX(MT_NORMAL_TAGGED);
>
> return __pgprot(prot);
> }
> EXPORT_SYMBOL(vm_get_page_prot);
>
> With that:
Sure, will change them into a single function.
>
> Reviewed-by: Catalin Marinas <[email protected]>
>