powerpc/64s/radix: Fix memory hotplug section page table creation
Nicholas Piggin
reviewerPATCH 1/5 powerpc/64s/radix: Fix memory hotplug section page table creation
create_physical_mapping expects physical addresses, but creating and splitting these mappings after boot is supplying virtual (effective) addresses. This can be irritated by booting with mem= to limit memory then probing an unused physical memory range:
echo <addr> > /sys/devices/system/memory/probe
This mostly works by accident, firstly because __va(__va(x)) == __va(x) so the virtual address does not get corrupted. Secondly because pfn_pte masks out the upper bits of the pfn beyond the physical address limit, so a pfn constructed with a 0xc000000000000000 virtual linear address will be masked back to the correct physical address in the pte.
--- arch/powerpc/mm/book3s64/radix_pgtable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/mm/book3s64/radix_pgtable.c +1 -1
@@ -902,7 +902,7 @@ int __meminit radix__create_section_mapping(unsigned long start, unsigned long ereturn -1 ; } -return create_physical_mapping (start, end, nid); +return create_physical_mapping (__pa (start),__pa (end), nid); }int __meminit radix__remove_section_mapping (unsigned long start,unsigned long end)
Signature · 1 line
2.22.0
Nicholas Piggin
reviewerPATCH 2/5 powerpc/64s/radix: Fix memory hot-unplug page table split
create_physical_mapping expects physical addresses, but splitting these mapping on hot unplug is supplying virtual (effective) addresses.
[I'm not sure how to test this one]
--- arch/powerpc/mm/book3s64/radix_pgtable.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/mm/book3s64/radix_pgtable.c +2 -2
@@ -737,8 +737,8 @@ static int __meminit stop_machine_change_mapping(void *data)spin_unlock (&init_mm.page_table_lock);pte_clear (&init_mm, params->aligned_start, params->pte); -create_physical_mapping (params->aligned_start, params->start,-1 ); -create_physical_mapping (params->end, params->aligned_end,-1 ); +create_physical_mapping (__pa (params->aligned_start),__pa (params->start),-1 ); +create_physical_mapping (__pa (params->end),__pa (params->aligned_end),-1 );spin_lock (&init_mm.page_table_lock);return 0 ; }
Signature · 1 line
2.22.0
Re: [PATCH 2/5] powerpc/64s/radix: Fix memory hot-unplug page table split
Quoted 6 lines · Nicholas Piggin
create_physical_mappingexpects physical addresses, but splitting these mapping on hot unplug is supplying virtual (effective) addresses.[I'm not sure how to test this one]
Memory hot unplug with kvm?
Quoted 24 lines
Cc Balbir Singh <[email protected]>Fixes 4dd5f8a99e791 ("powerpc/mm/radix: Split linear mapping on hot-unplug")Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/mm/book3s64/radix_pgtable.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)arch/powerpc/mm/book3s64/radix_pgtable.c +2
-2@@ -737,8 +737,8 @@ static int __meminit stop_machine_change_mapping(void *data)spin_unlock (&init_mm.page_table_lock);pte_clear (&init_mm, params->aligned_start, params->pte); -create_physical_mapping (params->aligned_start, params->start,-1 ); -create_physical_mapping (params->end, params->aligned_end,-1 ); +create_physical_mapping (__pa (params->aligned_start),__pa (params->start),-1 ); +create_physical_mapping (__pa (params->end),__pa (params->aligned_end),-1 );spin_lock (&init_mm.page_table_lock);return 0 ; }Signature · 1 line
2.22.0
Nicholas Piggin
reviewerPATCH 3/5 powerpc/perf: fix imc allocation failure handling
In reply to Nicholas PigginThe alloc_pages_node return value should be tested for failure before being passed to page_address.
--- arch/powerpc/perf/imc-pmu.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-)
arch/powerpc/perf/imc-pmu.c +18 -11
@@ -577,6 +577,7 @@ static int core_imc_mem_init(int cpu, int size) {int nid, rc =0 , core_id = (cpu / threads_per_core);struct imc_mem_info *mem_info; +struct page *page;/* * alloc_pages_node() will allocate memory for core in the
@@ -587,11 +588,12 @@ static int core_imc_mem_init(int cpu, int size) mem_info->id = core_id;/* We need only vbase for core counters */ - mem_info->vbase =page_address (alloc_pages_node (nid, -GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!mem_info->vbase) + page =alloc_pages_node (nid, +GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + mem_info->vbase =page_address (page);/* Init the mutex */ core_imc_refc[core_id].id = core_id;
@@ -849,15 +851,17 @@ static int thread_imc_mem_alloc(int cpu_id, int size)int nid =cpu_to_node (cpu_id);if (!local_mem) { +struct page *page;/* * This case could happen only once at start, since we dont * free the memory in cpu offline path. */ - local_mem =page_address (alloc_pages_node (nid, + page =alloc_pages_node (nid,GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!local_mem) + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + local_mem =page_address (page);per_cpu (thread_imc_mem, cpu_id) = local_mem; }
@@ -1095,11 +1099,14 @@ static int trace_imc_mem_alloc(int cpu_id, int size)int core_id = (cpu_id / threads_per_core);if (!local_mem) { - local_mem =page_address (alloc_pages_node (phys_id, -GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!local_mem) +struct page *page; + + page =alloc_pages_node (phys_id, +GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + local_mem =page_address (page);per_cpu (trace_imc_mem, cpu_id) = local_mem;/* Initialise the counters for trace mode */
Signature · 1 line
2.22.0
Re: [PATCH 3/5] powerpc/perf: fix imc allocation failure handling
Quoted 3 lines · Nicholas Piggin
The
alloc_pages_nodereturn value should be tested for failure before being passed topage_address.
This need Fixes: tag? It fix a real crash unlike the other patch in this series?
Quoted 78 lines
Cc Madhavan Srinivasan <[email protected]>Tested-by Anju T Sudhakar <[email protected]>Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/perf/imc-pmu.c | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-)arch/powerpc/perf/imc-pmu.c +18
-11@@ -577,6 +577,7 @@ static int core_imc_mem_init(int cpu, int size) {int nid, rc =0 , core_id = (cpu / threads_per_core);struct imc_mem_info *mem_info; +struct page *page;/* * alloc_pages_node() will allocate memory for core in the @@ -587,11 +588,12 @@ static int core_imc_mem_init(int cpu, int size) mem_info->id = core_id;/* We need only vbase for core counters */ - mem_info->vbase =page_address (alloc_pages_node (nid, -GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!mem_info->vbase) + page =alloc_pages_node (nid, +GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + mem_info->vbase =page_address (page);/* Init the mutex */ core_imc_refc[core_id].id = core_id;@@ -849,15 +851,17 @@ static int thread_imc_mem_alloc(int cpu_id, int size)int nid =cpu_to_node (cpu_id);if (!local_mem) { +struct page *page;/* * This case could happen only once at start, since we dont * free the memory in cpu offline path. */ - local_mem =page_address (alloc_pages_node (nid, + page =alloc_pages_node (nid,GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!local_mem) + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + local_mem =page_address (page);per_cpu (thread_imc_mem, cpu_id) = local_mem; }@@ -1095,11 +1099,14 @@ static int trace_imc_mem_alloc(int cpu_id, int size)int core_id = (cpu_id / threads_per_core);if (!local_mem) { - local_mem =page_address (alloc_pages_node (phys_id, -GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | - __GFP_NOWARN,get_order (size))); -if (!local_mem) +struct page *page; + + page =alloc_pages_node (phys_id, +GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE | + __GFP_NOWARN,get_order (size)); +if (!page)return -ENOMEM ; + local_mem =page_address (page);per_cpu (trace_imc_mem, cpu_id) = local_mem;/* Initialise the counters for trace mode */ Signature · 1 line
2.22.0
Nicholas Piggin
reviewerPATCH 4/5 powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
In reply to Nicholas PigginEnsure __va is given a physical address below PAGE_OFFSET, and __pa is given a virtual address above PAGE_OFFSET.
--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/page.h +12 -2
@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET * with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET)) -#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \ + (void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET); \ +}) + +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \ + (unsigned long)(x) & 0x0fffffffffffffffUL; \ +})#else /* 32-bit, non book E */ #define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) + PAGE_OFFSET - MEMORY_START))
Signature · 1 line
2.22.0
Re: [PATCH 4/5] powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
Quoted 3 lines · Nicholas Piggin
Ensure
__vais given a physical address belowPAGE_OFFSET, and__pais given a virtual address abovePAGE_OFFSET.
Quoted 12 lines
Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)arch/powerpc/include/asm/page.h +0
-0@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET
arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 224
Quoted 10 lines
arch/powerpc/include/asm/page.h +6
-2* with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET)) -#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \ + (void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET); \ +})
Can we make that static inline? Is there a need for __pa and __va to be #define like we do #ifndef anywhere?
Quoted 11 lines
arch/powerpc/include/asm/page.h +6
-0+ +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \ + (unsigned long)(x) & 0x0fffffffffffffffUL; \ +})#else /* 32-bit, non book E */ #define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) + PAGE_OFFSET - MEMORY_START))Signature · 1 line
2.22.0
Re: [PATCH 4/5] powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
In reply to Nicholas PigginLe 24/07/2019 à 10:46, Nicholas Piggin a écrit :
Quoted 13 lines
Ensure
__vais given a physical address belowPAGE_OFFSET, and__pais given a virtual address abovePAGE_OFFSET.Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)arch/powerpc/include/asm/page.h +0
-0
arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 222
Quoted 10 lines
arch/powerpc/include/asm/page.h +4
-2@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET * with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET)) -#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \
Do we really want to add a BUG_ON here ?
Can't we just add a WARN_ON, like in https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/arch/powerpc/include/asm/io.h?id=6bf752daca07c85c181159f75dcf65b12056883b ?
arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 228
Quoted 6 lines
arch/powerpc/include/asm/page.h +6
-0+ (void *)(unsigned long )((phys_addr_t)(x) |PAGE_OFFSET ); \ +}) + +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \
Same
arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 233
Quoted 6 lines
arch/powerpc/include/asm/page.h +2
-0+ (unsigned long )(x) &0x0fffffffffffffffUL ; \ +})#else /* 32-bit, non book E */ #define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) + PAGE_OFFSET - MEMORY_START))
Would it be possible to change those macros into static inlines ?
Christophe
Re: [PATCH 4/5] powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
In reply to Nicholas PigginHi Nic,
Le 24/07/2019 à 10:46, Nicholas Piggin a écrit :
Quoted 19 lines
Ensure
__vais given a physical address belowPAGE_OFFSET, and__pais given a virtual address abovePAGE_OFFSET.Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)arch/powerpc/include/asm/page.h +1
-1@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET * with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET))
arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 228
Quoted 10 lines
arch/powerpc/include/asm/page.h +9
-1-#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \ + (void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET); \ +}) + +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \
With this, it is likely that virt_addr_valid() BUGs on a non valid address.
I think the purpose of virt_addr_valid() is to check addresses seamlessly, see check_heap_object()
Quoted 5 lines
arch/powerpc/include/asm/page.h +2
-0+ (unsigned long )(x) &0x0fffffffffffffffUL ; \ +})#else /* 32-bit, non book E */ #define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) + PAGE_OFFSET - MEMORY_START))
Nicholas Piggin
reviewerRe: [PATCH 4/5] powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
Excerpts from Christophe Leroy's message of December 24, 2021 11:24 pm:
Quoted 35 lines
Hi Nic,
Le 24/07/2019 à 10:46, Nicholas Piggin a écrit :
Quoted 19 lines
Ensure
__vais given a physical address belowPAGE_OFFSET, and__pais given a virtual address abovePAGE_OFFSET.Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)arch/powerpc/include/asm/page.h +1
-1@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET * with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET))arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 228
Quoted 10 lines
arch/powerpc/include/asm/page.h +9
-1-#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \ + (void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET); \ +}) + +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \With this, it is likely that
virt_addr_valid()BUGs on a non valid address.
Quoted 2 lines
I think the purpose of
virt_addr_valid()is to check addresses seamlessly, seecheck_heap_object()
Looks like you're right. How did you catch that?
We could change virt_addr_valid() to make that test first. x86 and arm64 both do checking rather than relying on !pfn_valid for < PAGE_OFFSET addresses.
Thanks,
Nick
Re: [PATCH 4/5] powerpc/64: Add VIRTUAL_BUG_ON checks for __va and __pa addresses
Le 25/12/2021 à 11:10, Nicholas Piggin a écrit :
Quoted 36 lines
Excerpts from Christophe Leroy's message of December 24, 2021 11:24 pm:
Quoted 35 lines
Hi Nic,
Le 24/07/2019 à 10:46, Nicholas Piggin a écrit :
Quoted 19 lines
Ensure
__vais given a physical address belowPAGE_OFFSET, and__pais given a virtual address abovePAGE_OFFSET.Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/page.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)arch/powerpc/include/asm/page.h +1
-1@@ -215,9 +215,19 @@ static inline bool pfn_valid(unsigned long pfn)/* * gcc miscompiles (unsigned long)(&static_var) - PAGE_OFFSET * with -mcmodel=medium, so we use & and | instead of - and + on 64-bit. +* This also results in better code generation. */ -#define __va (x) ((void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET))arch/powerpc/include/asm/page.h static inline bool pfn_valid(unsigned long pfn) · line 228
Quoted 10 lines
arch/powerpc/include/asm/page.h +9
-1-#define __pa (x) ((unsigned long)(x) & 0x0fffffffffffffffUL) +#define __va (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) >= PAGE_OFFSET); \ + (void *)(unsigned long)((phys_addr_t)(x) | PAGE_OFFSET); \ +}) + +#define __pa (x) \ +({ \ + VIRTUAL_BUG_ON((unsigned long)(x) < PAGE_OFFSET); \With this, it is likely that
virt_addr_valid()BUGs on a non valid address.
Quoted 4 lines
Quoted 2 lines
I think the purpose of
virt_addr_valid()is to check addresses seamlessly, seecheck_heap_object()Looks like you're right. How did you catch that?
I caught that while looking at the problem reported by Kefeng where he says that virt_addr_valid() reports true on vmalloced memory on book3e/64
Quoted 4 lines
We could change
virt_addr_valid()to make that test first. x86 and arm64 both do checking rather than relying on !pfn_valid for <PAGE_OFFSETaddresses.
That should work.
Maybe also we should implement a __pa_nodebug() like x86 and arm64 ?
Christophe
Nicholas Piggin
reviewerPATCH 5/5 powerpc/64s/radix: Remove redundant pfn_pte bitop, add VM_BUG_ON
In reply to Nicholas Pigginpfn_pte is never given a pte above the addressable physical memory limit, so the masking is redundant. In case of a software bug, it is not obviously better to silently truncate the pfn than to corrupt the pte (either one will result in memory corruption or crashes), so there is no reason to add this to the fast path.
Add VM_BUG_ON to catch cases where the pfn is invalid. These would catch the create_section_mapping bug fixed by a previous commit.
[16885.256466] ------------[ cut here ]------------
[16885.256492] kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
cpu 0x0: Vector: 700 (Program Check) at [c0000000ee0a36d0]
pc: c000000000080738: __map_kernel_page+0x248/0x6f0
lr: c000000000080ac0: __map_kernel_page+0x5d0/0x6f0
sp: c0000000ee0a3960
msr: 9000000000029033
current = 0xc0000000ec63b400
paca = 0xc0000000017f0000 irqmask: 0x03 irq_happened: 0x01
pid = 85, comm = sh
kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
Linux version 5.3.0-rc1-00001-g0fe93e5f3394
enter ? for help
[c0000000ee0a3a00] c000000000d37378 create_physical_mapping+0x260/0x360
[c0000000ee0a3b10] c000000000d370bc create_section_mapping+0x1c/0x3c
[c0000000ee0a3b30] c000000000071f54 arch_add_memory+0x74/0x130--- arch/powerpc/include/asm/book3s/64/pgtable.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/book3s/64/pgtable.h +4 -2
@@ -608,8 +608,10 @@ static inline bool pte_access_permitted(pte_t pte, bool write)*/ static inline pte_t pfn_pte (unsigned long pfn,pgprot_t pgprot) { -return __pte ((((pte_basic_t)(pfn) <<PAGE_SHIFT ) &PTE_RPN_MASK ) | -pgprot_val (pgprot)); +VM_BUG_ON (pfn >> (64 -PAGE_SHIFT )); +VM_BUG_ON ((pfn <<PAGE_SHIFT ) & ~PTE_RPN_MASK ); + +return __pte (((pte_basic_t )pfn <<PAGE_SHIFT ) |pgprot_val (pgprot)); }static inline unsigned long pte_pfn (pte_t pte)
Signature · 1 line
2.22.0
Re: [PATCH 5/5] powerpc/64s/radix: Remove redundant pfn_pte bitop, add VM_BUG_ON
Quoted 9 lines · Nicholas Piggin
pfn_pteis never given a pte above the addressable physical memory limit, so the masking is redundant. In case of a software bug, it is not obviously better to silently truncate the pfn than to corrupt the pte (either one will result in memory corruption or crashes), so there is no reason to add this to the fast path.Add
VM_BUG_ONto catch cases where the pfn is invalid. These would catch thecreate_section_mappingbug fixed by a previous commit.
Quoted 17 lines
[16885.256466] ------------[ cut here ]------------
[16885.256492] kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
cpu 0x0: Vector: 700 (Program Check) at [c0000000ee0a36d0]
pc: c000000000080738: __map_kernel_page+0x248/0x6f0
lr: c000000000080ac0: __map_kernel_page+0x5d0/0x6f0
sp: c0000000ee0a3960
msr: 9000000000029033
current = 0xc0000000ec63b400
paca = 0xc0000000017f0000 irqmask: 0x03 irq_happened: 0x01
pid = 85, comm = sh
kernel BUG at arch/powerpc/include/asm/book3s/64/pgtable.h:612!
Linux version 5.3.0-rc1-00001-g0fe93e5f3394
enter ? for help
[c0000000ee0a3a00] c000000000d37378 create_physical_mapping+0x260/0x360
[c0000000ee0a3b10] c000000000d370bc create_section_mapping+0x1c/0x3c
[c0000000ee0a3b30] c000000000071f54 arch_add_memory+0x74/0x130Quoted 24 lines
Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/include/asm/book3s/64/pgtable.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)arch/powerpc/include/asm/book3s/64/pgtable.h +4
-2@@ -608,8 +608,10 @@ static inline bool pte_access_permitted(pte_t pte, bool write)*/ static inline pte_t pfn_pte (unsigned long pfn,pgprot_t pgprot) { -return __pte ((((pte_basic_t)(pfn) <<PAGE_SHIFT ) &PTE_RPN_MASK ) | -pgprot_val (pgprot)); +VM_BUG_ON (pfn >> (64 -PAGE_SHIFT )); +VM_BUG_ON ((pfn <<PAGE_SHIFT ) & ~PTE_RPN_MASK ); + +return __pte (((pte_basic_t )pfn <<PAGE_SHIFT ) |pgprot_val (pgprot)); }static inline unsigned long pte_pfn (pte_t pte)Signature · 1 line
2.22.0
Quoted 5 lines · Nicholas Piggin
create_physical_mappingexpects physical addresses, but creating and splitting these mappings after boot is supplying virtual (effective) addresses. This can be irritated by booting with mem= to limit memory then probing an unused physical memory range:
Quoted 8 lines
echo <addr> >
/sys/devices/system/memory/probeThis mostly works by accident, firstly because __va(__va(x)) ==
__va(x)so the virtual address does not get corrupted. Secondly becausepfn_ptemasks out the upper bits of the pfn beyond the physical address limit, so a pfn constructed with a0xc000000000000000virtual linear address will be masked back to the correct physical address in the pte.
Good catch. Did this result in any error?
Quoted 12 lines
Cc Reza Arbab <[email protected]>Fixes 6cc27341b21a8 ("powerpc/mm: add radix__create_section_mapping()")Signed-off-by Nicholas Piggin <[email protected]>--- arch/powerpc/mm/book3s64/radix_pgtable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)arch/powerpc/mm/book3s64/radix_pgtable.c +0
-0
arch/powerpc/mm/book3s64/radix_pgtable.c int __meminit radix__create_section_mapping(unsigned long start, unsigned long e · line 905
Quoted 5 lines
arch/powerpc/mm/book3s64/radix_pgtable.c +1
-1@@ -902,7 +902,7 @@ int __meminit radix__create_section_mapping(unsigned long start, unsigned long ereturn -1 ; } -return create_physical_mapping (start, end, nid); +return create_physical_mapping (__pa (start),__pa (end), nid);
While we are here, should we change the prototype to take phys_addr_t ?
Quoted 5 lines
arch/powerpc/mm/book3s64/radix_pgtable.c +0
-0}int __meminit radix__remove_section_mapping (unsigned long start,unsigned long end)Signature · 1 line
2.22.0
Quoted 13 lines · Nicholas Piggin
create_physical_mappingexpects physical addresses, but creating and splitting these mappings after boot is supplying virtual (effective) addresses. This can be irritated by booting with mem= to limit memory then probing an unused physical memory range:echo <addr> >
/sys/devices/system/memory/probeThis mostly works by accident, firstly because __va(__va(x)) ==
__va(x)so the virtual address does not get corrupted. Secondly becausepfn_ptemasks out the upper bits of the pfn beyond the physical address limit, so a pfn constructed with a0xc000000000000000virtual linear address will be masked back to the correct physical address in the pte.
Quoted 4 lines
Cc Reza Arbab <[email protected]>Fixes 6cc27341b21a8 ("powerpc/mm: add radix__create_section_mapping()")Signed-off-by Nicholas Piggin <[email protected]>Reviewed-by Aneesh Kumar K.V <[email protected]>
Series applied to powerpc next, thanks.
https://git.kernel.org/powerpc/c/8f51e3929470942e6a8744061254fdeef646cd36
cheers