Commit af73e4d9 authored by Naoya Horiguchi's avatar Naoya Horiguchi Committed by Linus Torvalds

hugetlbfs: fix mmap failure in unaligned size request

The current kernel returns -EINVAL unless a given mmap length is
"almost" hugepage aligned.  This is because in sys_mmap_pgoff() the
given length is passed to vm_mmap_pgoff() as it is without being aligned
with hugepage boundary.

This is a regression introduced in commit 40716e29 ("hugetlbfs: fix
alignment of huge page requests"), where alignment code is pushed into
hugetlb_file_setup() and the variable len in caller side is not changed.

To fix this, this patch partially reverts that commit, and adds
alignment code in caller side.  And it also introduces hstate_sizelog()
in order to get proper hstate to specified hugepage size.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=56881

[akpm@linux-foundation.org: fix warning when CONFIG_HUGETLB_PAGE=n]
Signed-off-by: default avatarNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: default avatarJohannes Weiner <hannes@cmpxchg.org>
Reported-by: <iceman_dvd@yahoo.com>
Cc: Steven Truelove <steven.truelove@utoronto.ca>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 1ab4ce76
...@@ -909,11 +909,8 @@ static int can_do_hugetlb_shm(void) ...@@ -909,11 +909,8 @@ static int can_do_hugetlb_shm(void)
static int get_hstate_idx(int page_size_log) static int get_hstate_idx(int page_size_log)
{ {
struct hstate *h; struct hstate *h = hstate_sizelog(page_size_log);
if (!page_size_log)
return default_hstate_idx;
h = size_to_hstate(1 << page_size_log);
if (!h) if (!h)
return -1; return -1;
return h - hstates; return h - hstates;
...@@ -929,9 +926,12 @@ static struct dentry_operations anon_ops = { ...@@ -929,9 +926,12 @@ static struct dentry_operations anon_ops = {
.d_dname = hugetlb_dname .d_dname = hugetlb_dname
}; };
struct file *hugetlb_file_setup(const char *name, unsigned long addr, /*
size_t size, vm_flags_t acctflag, * Note that size should be aligned to proper hugepage size in caller side,
struct user_struct **user, * otherwise hugetlb_reserve_pages reserves one less hugepages than intended.
*/
struct file *hugetlb_file_setup(const char *name, size_t size,
vm_flags_t acctflag, struct user_struct **user,
int creat_flags, int page_size_log) int creat_flags, int page_size_log)
{ {
struct file *file = ERR_PTR(-ENOMEM); struct file *file = ERR_PTR(-ENOMEM);
...@@ -939,8 +939,6 @@ struct file *hugetlb_file_setup(const char *name, unsigned long addr, ...@@ -939,8 +939,6 @@ struct file *hugetlb_file_setup(const char *name, unsigned long addr,
struct path path; struct path path;
struct super_block *sb; struct super_block *sb;
struct qstr quick_string; struct qstr quick_string;
struct hstate *hstate;
unsigned long num_pages;
int hstate_idx; int hstate_idx;
hstate_idx = get_hstate_idx(page_size_log); hstate_idx = get_hstate_idx(page_size_log);
...@@ -980,12 +978,10 @@ struct file *hugetlb_file_setup(const char *name, unsigned long addr, ...@@ -980,12 +978,10 @@ struct file *hugetlb_file_setup(const char *name, unsigned long addr,
if (!inode) if (!inode)
goto out_dentry; goto out_dentry;
hstate = hstate_inode(inode);
size += addr & ~huge_page_mask(hstate);
num_pages = ALIGN(size, huge_page_size(hstate)) >>
huge_page_shift(hstate);
file = ERR_PTR(-ENOMEM); file = ERR_PTR(-ENOMEM);
if (hugetlb_reserve_pages(inode, 0, num_pages, NULL, acctflag)) if (hugetlb_reserve_pages(inode, 0,
size >> huge_page_shift(hstate_inode(inode)), NULL,
acctflag))
goto out_inode; goto out_inode;
d_instantiate(path.dentry, inode); d_instantiate(path.dentry, inode);
......
...@@ -189,8 +189,7 @@ static inline struct hugetlbfs_sb_info *HUGETLBFS_SB(struct super_block *sb) ...@@ -189,8 +189,7 @@ static inline struct hugetlbfs_sb_info *HUGETLBFS_SB(struct super_block *sb)
extern const struct file_operations hugetlbfs_file_operations; extern const struct file_operations hugetlbfs_file_operations;
extern const struct vm_operations_struct hugetlb_vm_ops; extern const struct vm_operations_struct hugetlb_vm_ops;
struct file *hugetlb_file_setup(const char *name, unsigned long addr, struct file *hugetlb_file_setup(const char *name, size_t size, vm_flags_t acct,
size_t size, vm_flags_t acct,
struct user_struct **user, int creat_flags, struct user_struct **user, int creat_flags,
int page_size_log); int page_size_log);
...@@ -209,8 +208,8 @@ static inline int is_file_hugepages(struct file *file) ...@@ -209,8 +208,8 @@ static inline int is_file_hugepages(struct file *file)
#define is_file_hugepages(file) 0 #define is_file_hugepages(file) 0
static inline struct file * static inline struct file *
hugetlb_file_setup(const char *name, unsigned long addr, size_t size, hugetlb_file_setup(const char *name, size_t size, vm_flags_t acctflag,
vm_flags_t acctflag, struct user_struct **user, int creat_flags, struct user_struct **user, int creat_flags,
int page_size_log) int page_size_log)
{ {
return ERR_PTR(-ENOSYS); return ERR_PTR(-ENOSYS);
...@@ -288,6 +287,13 @@ static inline struct hstate *hstate_file(struct file *f) ...@@ -288,6 +287,13 @@ static inline struct hstate *hstate_file(struct file *f)
return hstate_inode(file_inode(f)); return hstate_inode(file_inode(f));
} }
static inline struct hstate *hstate_sizelog(int page_size_log)
{
if (!page_size_log)
return &default_hstate;
return size_to_hstate(1 << page_size_log);
}
static inline struct hstate *hstate_vma(struct vm_area_struct *vma) static inline struct hstate *hstate_vma(struct vm_area_struct *vma)
{ {
return hstate_file(vma->vm_file); return hstate_file(vma->vm_file);
...@@ -352,11 +358,12 @@ static inline int hstate_index(struct hstate *h) ...@@ -352,11 +358,12 @@ static inline int hstate_index(struct hstate *h)
return h - hstates; return h - hstates;
} }
#else #else /* CONFIG_HUGETLB_PAGE */
struct hstate {}; struct hstate {};
#define alloc_huge_page_node(h, nid) NULL #define alloc_huge_page_node(h, nid) NULL
#define alloc_bootmem_huge_page(h) NULL #define alloc_bootmem_huge_page(h) NULL
#define hstate_file(f) NULL #define hstate_file(f) NULL
#define hstate_sizelog(s) NULL
#define hstate_vma(v) NULL #define hstate_vma(v) NULL
#define hstate_inode(i) NULL #define hstate_inode(i) NULL
#define huge_page_size(h) PAGE_SIZE #define huge_page_size(h) PAGE_SIZE
...@@ -371,6 +378,6 @@ static inline unsigned int pages_per_huge_page(struct hstate *h) ...@@ -371,6 +378,6 @@ static inline unsigned int pages_per_huge_page(struct hstate *h)
} }
#define hstate_index_to_shift(index) 0 #define hstate_index_to_shift(index) 0
#define hstate_index(h) 0 #define hstate_index(h) 0
#endif #endif /* CONFIG_HUGETLB_PAGE */
#endif /* _LINUX_HUGETLB_H */ #endif /* _LINUX_HUGETLB_H */
...@@ -491,10 +491,14 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) ...@@ -491,10 +491,14 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
sprintf (name, "SYSV%08x", key); sprintf (name, "SYSV%08x", key);
if (shmflg & SHM_HUGETLB) { if (shmflg & SHM_HUGETLB) {
struct hstate *hs = hstate_sizelog((shmflg >> SHM_HUGE_SHIFT)
& SHM_HUGE_MASK);
size_t hugesize = ALIGN(size, huge_page_size(hs));
/* hugetlb_file_setup applies strict accounting */ /* hugetlb_file_setup applies strict accounting */
if (shmflg & SHM_NORESERVE) if (shmflg & SHM_NORESERVE)
acctflag = VM_NORESERVE; acctflag = VM_NORESERVE;
file = hugetlb_file_setup(name, 0, size, acctflag, file = hugetlb_file_setup(name, hugesize, acctflag,
&shp->mlock_user, HUGETLB_SHMFS_INODE, &shp->mlock_user, HUGETLB_SHMFS_INODE,
(shmflg >> SHM_HUGE_SHIFT) & SHM_HUGE_MASK); (shmflg >> SHM_HUGE_SHIFT) & SHM_HUGE_MASK);
} else { } else {
......
...@@ -1363,15 +1363,20 @@ SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len, ...@@ -1363,15 +1363,20 @@ SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len,
file = fget(fd); file = fget(fd);
if (!file) if (!file)
goto out; goto out;
if (is_file_hugepages(file))
len = ALIGN(len, huge_page_size(hstate_file(file)));
} else if (flags & MAP_HUGETLB) { } else if (flags & MAP_HUGETLB) {
struct user_struct *user = NULL; struct user_struct *user = NULL;
len = ALIGN(len, huge_page_size(hstate_sizelog(
(flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK)));
/* /*
* VM_NORESERVE is used because the reservations will be * VM_NORESERVE is used because the reservations will be
* taken when vm_ops->mmap() is called * taken when vm_ops->mmap() is called
* A dummy user value is used because we are not locking * A dummy user value is used because we are not locking
* memory so no accounting is necessary * memory so no accounting is necessary
*/ */
file = hugetlb_file_setup(HUGETLB_ANON_FILE, addr, len, file = hugetlb_file_setup(HUGETLB_ANON_FILE, len,
VM_NORESERVE, VM_NORESERVE,
&user, HUGETLB_ANONHUGE_INODE, &user, HUGETLB_ANONHUGE_INODE,
(flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK); (flags >> MAP_HUGE_SHIFT) & MAP_HUGE_MASK);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment