From e130242dc351f1cfa2bbeb6766a1486ce936ef88 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Sep 2021 15:18:21 -0700 Subject: mm: simplify compat numa syscalls The compat implementations for mbind, get_mempolicy, set_mempolicy and migrate_pages are just there to handle the subtly different layout of bitmaps on 32-bit hosts. The compat implementation however lacks some of the checks that are present in the native one, in particular for checking that the extra bits are all zero when user space has a larger mask size than the kernel. Worse, those extra bits do not get cleared when copying in or out of the kernel, which can lead to incorrect data as well. Unify the implementation to handle the compat bitmap layout directly in the get_nodes() and copy_nodes_to_user() helpers. Splitting out the get_bitmap() helper from get_nodes() also helps readability of the native case. On x86, two additional problems are addressed by this: compat tasks can pass a bitmap at the end of a mapping, causing a fault when reading across the page boundary for a 64-bit word. x32 tasks might also run into problems with get_mempolicy corrupting data when an odd number of 32-bit words gets passed. On parisc the migrate_pages() system call apparently had the wrong calling convention, as big-endian architectures expect the words inside of a bitmap to be swapped. This is not a problem though since parisc has no NUMA support. [arnd@arndb.de: fix mempolicy crash] Link: https://lkml.kernel.org/r/20210730143417.3700653-1-arnd@kernel.org Link: https://lore.kernel.org/lkml/YQPLG20V3dmOfq3a@osiris/ Link: https://lkml.kernel.org/r/20210727144859.4150043-5-arnd@kernel.org Signed-off-by: Arnd Bergmann Reviewed-by: Christoph Hellwig Cc: Al Viro Cc: Benjamin Herrenschmidt Cc: Borislav Petkov Cc: Catalin Marinas Cc: Christian Borntraeger Cc: Christoph Hellwig Cc: "David S. Miller" Cc: Eric Biederman Cc: Feng Tang Cc: Heiko Carstens Cc: Helge Deller Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Michael Ellerman Cc: Paul Mackerras Cc: Thomas Bogendoerfer Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compat.h b/include/linux/compat.h index 8e0598c7d1d1..3a2ac5afee30 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -395,14 +395,6 @@ struct compat_kexec_segment; struct compat_mq_attr; struct compat_msgbuf; -#define BITS_PER_COMPAT_LONG (8*sizeof(compat_long_t)) - -#define BITS_TO_COMPAT_LONGS(bits) DIV_ROUND_UP(bits, BITS_PER_COMPAT_LONG) - -long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, - unsigned long bitmap_size); -long compat_put_bitmap(compat_ulong_t __user *umask, unsigned long *mask, - unsigned long bitmap_size); void copy_siginfo_to_external32(struct compat_siginfo *to, const struct kernel_siginfo *from); int copy_siginfo_from_user32(kernel_siginfo_t *to, @@ -976,6 +968,15 @@ static inline bool in_compat_syscall(void) { return false; } #endif /* CONFIG_COMPAT */ +#define BITS_PER_COMPAT_LONG (8*sizeof(compat_long_t)) + +#define BITS_TO_COMPAT_LONGS(bits) DIV_ROUND_UP(bits, BITS_PER_COMPAT_LONG) + +long compat_get_bitmap(unsigned long *mask, const compat_ulong_t __user *umask, + unsigned long bitmap_size); +long compat_put_bitmap(compat_ulong_t __user *umask, unsigned long *mask, + unsigned long bitmap_size); + /* * Some legacy ABIs like the i386 one use less than natural alignment for 64-bit * types, and will need special compat treatment for that. Most architectures -- cgit From 59ab844eed9c6b01d32dcb27b57accc23771b324 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Sep 2021 15:18:25 -0700 Subject: compat: remove some compat entry points These are all handled correctly when calling the native system call entry point, so remove the special cases. Link: https://lkml.kernel.org/r/20210727144859.4150043-6-arnd@kernel.org Signed-off-by: Arnd Bergmann Reviewed-by: Christoph Hellwig Cc: Al Viro Cc: Benjamin Herrenschmidt Cc: Borislav Petkov Cc: Catalin Marinas Cc: Christian Borntraeger Cc: Christoph Hellwig Cc: "David S. Miller" Cc: Eric Biederman Cc: Feng Tang Cc: Heiko Carstens Cc: Helge Deller Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Michael Ellerman Cc: Paul Mackerras Cc: Thomas Bogendoerfer Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 20 -------------------- 1 file changed, 20 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compat.h b/include/linux/compat.h index 3a2ac5afee30..2d42cebd1fb8 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -799,26 +799,6 @@ asmlinkage long compat_sys_execve(const char __user *filename, const compat_uptr /* mm/fadvise.c: No generic prototype for fadvise64_64 */ /* mm/, CONFIG_MMU only */ -asmlinkage long compat_sys_mbind(compat_ulong_t start, compat_ulong_t len, - compat_ulong_t mode, - compat_ulong_t __user *nmask, - compat_ulong_t maxnode, compat_ulong_t flags); -asmlinkage long compat_sys_get_mempolicy(int __user *policy, - compat_ulong_t __user *nmask, - compat_ulong_t maxnode, - compat_ulong_t addr, - compat_ulong_t flags); -asmlinkage long compat_sys_set_mempolicy(int mode, compat_ulong_t __user *nmask, - compat_ulong_t maxnode); -asmlinkage long compat_sys_migrate_pages(compat_pid_t pid, - compat_ulong_t maxnode, const compat_ulong_t __user *old_nodes, - const compat_ulong_t __user *new_nodes); -asmlinkage long compat_sys_move_pages(pid_t pid, compat_ulong_t nr_pages, - __u32 __user *pages, - const int __user *nodes, - int __user *status, - int flags); - asmlinkage long compat_sys_rt_tgsigqueueinfo(compat_pid_t tgid, compat_pid_t pid, int sig, struct compat_siginfo __user *uinfo); -- cgit From a7a08b275a8bbade798c4bdaad07ade68fe7003c Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 8 Sep 2021 15:18:29 -0700 Subject: arch: remove compat_alloc_user_space All users of compat_alloc_user_space() and copy_in_user() have been removed from the kernel, only a few functions in sparc remain that can be changed to calling arch_copy_in_user() instead. Link: https://lkml.kernel.org/r/20210727144859.4150043-7-arnd@kernel.org Signed-off-by: Arnd Bergmann Reviewed-by: Christoph Hellwig Cc: Al Viro Cc: Benjamin Herrenschmidt Cc: Borislav Petkov Cc: Catalin Marinas Cc: Christian Borntraeger Cc: Christoph Hellwig Cc: "David S. Miller" Cc: Eric Biederman Cc: Feng Tang Cc: Heiko Carstens Cc: Helge Deller Cc: "H. Peter Anvin" Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Michael Ellerman Cc: Paul Mackerras Cc: Thomas Bogendoerfer Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/compat.h | 2 -- include/linux/uaccess.h | 10 ---------- 2 files changed, 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compat.h b/include/linux/compat.h index 2d42cebd1fb8..1c758b0e0359 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -511,8 +511,6 @@ extern long compat_arch_ptrace(struct task_struct *child, compat_long_t request, struct epoll_event; /* fortunately, this one is fixed-layout */ -extern void __user *compat_alloc_user_space(unsigned long len); - int compat_restore_altstack(const compat_stack_t __user *uss); int __compat_save_altstack(compat_stack_t __user *, unsigned long); #define unsafe_compat_save_altstack(uss, sp, label) do { \ diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index c05e903cef02..ac0394087f7d 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -200,16 +200,6 @@ copy_to_user(void __user *to, const void *from, unsigned long n) n = _copy_to_user(to, from, n); return n; } -#ifdef CONFIG_COMPAT -static __always_inline unsigned long __must_check -copy_in_user(void __user *to, const void __user *from, unsigned long n) -{ - might_fault(); - if (access_ok(to, n) && access_ok(from, n)) - n = raw_copy_in_user(to, from, n); - return n; -} -#endif #ifndef copy_mc_to_kernel /* -- cgit From 13db8c50477d83ad3e3b9b0ae247e5cd833a7ae4 Mon Sep 17 00:00:00 2001 From: Liu Zixian Date: Wed, 8 Sep 2021 18:10:05 -0700 Subject: mm/hugetlb: initialize hugetlb_usage in mm_init After fork, the child process will get incorrect (2x) hugetlb_usage. If a process uses 5 2MB hugetlb pages in an anonymous mapping, HugetlbPages: 10240 kB and then forks, the child will show, HugetlbPages: 20480 kB The reason for double the amount is because hugetlb_usage will be copied from the parent and then increased when we copy page tables from parent to child. Child will have 2x actual usage. Fix this by adding hugetlb_count_init in mm_init. Link: https://lkml.kernel.org/r/20210826071742.877-1-liuzixian4@huawei.com Fixes: 5d317b2b6536 ("mm: hugetlb: proc: add HugetlbPages field to /proc/PID/status") Signed-off-by: Liu Zixian Reviewed-by: Naoya Horiguchi Reviewed-by: Mike Kravetz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index f7ca1a3870ea..1faebe1cd0ed 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -858,6 +858,11 @@ static inline spinlock_t *huge_pte_lockptr(struct hstate *h, void hugetlb_report_usage(struct seq_file *m, struct mm_struct *mm); +static inline void hugetlb_count_init(struct mm_struct *mm) +{ + atomic_long_set(&mm->hugetlb_usage, 0); +} + static inline void hugetlb_count_add(long l, struct mm_struct *mm) { atomic_long_add(l, &mm->hugetlb_usage); @@ -1042,6 +1047,10 @@ static inline spinlock_t *huge_pte_lockptr(struct hstate *h, return &mm->page_table_lock; } +static inline void hugetlb_count_init(struct mm_struct *mm) +{ +} + static inline void hugetlb_report_usage(struct seq_file *f, struct mm_struct *m) { } -- cgit From 10994316089c9682f2fbe0be0b1e82bcaf5f4e8c Mon Sep 17 00:00:00 2001 From: Liam Howlett Date: Wed, 8 Sep 2021 18:10:14 -0700 Subject: mmap_lock: change trace and locking order Print to the trace log before releasing the lock to avoid racing with other trace log printers of the same lock type. Link: https://lkml.kernel.org/r/20210903022041.1843024-1-Liam.Howlett@oracle.com Signed-off-by: Liam R. Howlett Suggested-by: Steven Rostedt (VMware) Reviewed-by: Matthew Wilcox (Oracle) Cc: Michel Lespinasse Cc: Vlastimil Babka Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmap_lock.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h index 0540f0156f58..b179f1e3541a 100644 --- a/include/linux/mmap_lock.h +++ b/include/linux/mmap_lock.h @@ -101,14 +101,14 @@ static inline bool mmap_write_trylock(struct mm_struct *mm) static inline void mmap_write_unlock(struct mm_struct *mm) { - up_write(&mm->mmap_lock); __mmap_lock_trace_released(mm, true); + up_write(&mm->mmap_lock); } static inline void mmap_write_downgrade(struct mm_struct *mm) { - downgrade_write(&mm->mmap_lock); __mmap_lock_trace_acquire_returned(mm, false, true); + downgrade_write(&mm->mmap_lock); } static inline void mmap_read_lock(struct mm_struct *mm) @@ -140,8 +140,8 @@ static inline bool mmap_read_trylock(struct mm_struct *mm) static inline void mmap_read_unlock(struct mm_struct *mm) { - up_read(&mm->mmap_lock); __mmap_lock_trace_released(mm, false); + up_read(&mm->mmap_lock); } static inline bool mmap_read_trylock_non_owner(struct mm_struct *mm) @@ -155,8 +155,8 @@ static inline bool mmap_read_trylock_non_owner(struct mm_struct *mm) static inline void mmap_read_unlock_non_owner(struct mm_struct *mm) { - up_read_non_owner(&mm->mmap_lock); __mmap_lock_trace_released(mm, false); + up_read_non_owner(&mm->mmap_lock); } static inline void mmap_assert_locked(struct mm_struct *mm) -- cgit