diff options
| author | Lorenzo Stoakes <ljs@kernel.org> | 2026-05-12 17:06:43 +0100 |
|---|---|---|
| committer | Andrew Morton <akpm@linux-foundation.org> | 2026-05-21 19:06:11 -0700 |
| commit | 83f9efcce93f8574be2279090ee2aec58b86cda7 (patch) | |
| tree | 785daedb15ce5bdb1d6e5dd5188f54bef60ba8d4 /fs | |
| parent | 83ec6eeb74a592e6568cb0723bac99fb8b3810b4 (diff) | |
| download | linux-next-history-83f9efcce93f8574be2279090ee2aec58b86cda7.tar.gz | |
Revert "mm/hugetlbfs: update hugetlbfs to use mmap_prepare"
This reverts commit ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use
mmap_prepare") with conflict resolution to account for changes in commit
ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use mmap_prepare").
The patch incorrectly handled hugetlb VMA lock allocation at the
mmap_prepare stage, where a failed allocation occurring after mmap_prepare
is called might result in the lock leaking.
There is no risk of a merge causing a similar issues, as
VMA_DONTEXPAND_BIT is set for hugetlb mappings.
As a first step in addressing this issue, simply revert the change so we
can rework how we do this having corrected the underlying issues.
We maintain the VMA flags changes as best we can, accounting for the fact
that we were working with a VMA descriptor previously and propagating
like-for-like changes for this.
Note that we invoke vma_set_flags() and do not call vma_start_write() as
vm_flags_set() does. This is OK as it's being done in an .mmap hook where
the VMA is not yet linked into the tree so nobody else can be accessing
it.
Link: https://lore.kernel.org/20260512160643.266960-1-ljs@kernel.org
Fixes: ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use mmap_prepare")
Signed-off-by: Lorenzo Stoakes <ljs@kernel.org>
Reported-by: Mingyu Wang <25181214217@stu.xidian.edu.cn>
Closes: https://lore.kernel.org/linux-mm/20260425070700.562229-1-25181214217@stu.xidian.edu.cn/
Acked-by: Muchun Song <muchun.song@linux.dev>
Acked-by: Oscar Salvador <osalvador@suse.de>
Cc: David Hildenbrand <david@kernel.org>
Cc: Liam R. Howlett <liam@infradead.org>
Cc: Pedro Falcato <pfalcato@suse.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/hugetlbfs/inode.c | 46 |
1 files changed, 13 insertions, 33 deletions
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 8b05bec08e049..78d61bf2bd9bb 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -96,15 +96,8 @@ static const struct fs_parameter_spec hugetlb_fs_parameters[] = { #define PGOFF_LOFFT_MAX \ (((1UL << (PAGE_SHIFT + 1)) - 1) << (BITS_PER_LONG - (PAGE_SHIFT + 1))) -static int hugetlb_file_mmap_prepare_success(const struct vm_area_struct *vma) +static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) { - /* Unfortunate we have to reassign vma->vm_private_data. */ - return hugetlb_vma_lock_alloc((struct vm_area_struct *)vma); -} - -static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) -{ - struct file *file = desc->file; struct inode *inode = file_inode(file); loff_t len, vma_len; int ret; @@ -119,8 +112,8 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) * way when do_mmap unwinds (may be important on powerpc * and ia64). */ - vma_desc_set_flags(desc, VMA_HUGETLB_BIT, VMA_DONTEXPAND_BIT); - desc->vm_ops = &hugetlb_vm_ops; + vma_set_flags(vma, VMA_HUGETLB_BIT, VMA_DONTEXPAND_BIT); + vma->vm_ops = &hugetlb_vm_ops; /* * page based offset in vm_pgoff could be sufficiently large to @@ -129,16 +122,16 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) * sizeof(unsigned long). So, only check in those instances. */ if (sizeof(unsigned long) == sizeof(loff_t)) { - if (desc->pgoff & PGOFF_LOFFT_MAX) + if (vma->vm_pgoff & PGOFF_LOFFT_MAX) return -EINVAL; } /* must be huge page aligned */ - if (desc->pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) + if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) return -EINVAL; - vma_len = (loff_t)vma_desc_size(desc); - len = vma_len + ((loff_t)desc->pgoff << PAGE_SHIFT); + vma_len = (loff_t)(vma->vm_end - vma->vm_start); + len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); /* check for overflow */ if (len < vma_len) return -EINVAL; @@ -148,7 +141,7 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) ret = -ENOMEM; - vma_flags = desc->vma_flags; + vma_flags = vma->flags; /* * for SHM_HUGETLB, the pages are reserved in the shmget() call so skip * reserving here. Note: only for SHM hugetlbfs file, the inode @@ -158,30 +151,17 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) vma_flags_set(&vma_flags, VMA_NORESERVE_BIT); if (hugetlb_reserve_pages(inode, - desc->pgoff >> huge_page_order(h), - len >> huge_page_shift(h), desc, - vma_flags) < 0) + vma->vm_pgoff >> huge_page_order(h), + len >> huge_page_shift(h), vma, + vma_flags) < 0) goto out; ret = 0; - if (vma_desc_test(desc, VMA_WRITE_BIT) && inode->i_size < len) + if (vma_test(vma, VMA_WRITE_BIT) && inode->i_size < len) i_size_write(inode, len); out: inode_unlock(inode); - if (!ret) { - /* Allocate the VMA lock after we set it up. */ - desc->action.success_hook = hugetlb_file_mmap_prepare_success; - /* - * We cannot permit the rmap finding this VMA in the time - * between the VMA being inserted into the VMA tree and the - * completion/success hook being invoked. - * - * This is because we establish a per-VMA hugetlb lock which can - * be raced by rmap. - */ - desc->action.hide_from_rmap_until_complete = true; - } return ret; } @@ -1227,7 +1207,7 @@ static void init_once(void *foo) static const struct file_operations hugetlbfs_file_operations = { .read_iter = hugetlbfs_read_iter, - .mmap_prepare = hugetlbfs_file_mmap_prepare, + .mmap = hugetlbfs_file_mmap, .fsync = noop_fsync, .get_unmapped_area = hugetlb_get_unmapped_area, .llseek = default_llseek, |