On Sun, Dec 31, 2023 at 12:15:54PM -0800, Darrick J. Wong wrote:
> +/* Ensure that there is storage backing the given range. */
> +int
> +xfile_prealloc(
> + struct xfile *xf,
> + loff_t pos,
> + u64 count)
> +{
> + struct inode *inode = file_inode(xf->file);
> + struct address_space *mapping = inode->i_mapping;
> + const struct address_space_operations *aops = mapping->a_ops;
> + struct page *page = NULL;
> + unsigned int pflags;
> + int error = 0;
> +
> + if (count > MAX_RW_COUNT)
> + return -E2BIG;
> + if (inode->i_sb->s_maxbytes - pos < count)
> + return -EFBIG;
> +
> + trace_xfile_prealloc(xf, pos, count);
> +
> + pflags = memalloc_nofs_save();
> + while (count > 0) {
> + void *fsdata = NULL;
> + unsigned int len;
> + int ret;
> +
> + len = min_t(ssize_t, count, PAGE_SIZE - offset_in_page(pos));
> +
> + /*
> + * We call write_begin directly here to avoid all the freezer
> + * protection lock-taking that happens in the normal path.
> + * shmem doesn't support fs freeze, but lockdep doesn't know
> + * that and will trip over that.
> + */
> + error = aops->write_begin(NULL, mapping, pos, len, &page,
> + &fsdata);
> + if (error)
> + break;
> +
> + /*
> + * xfile pages must never be mapped into userspace, so we skip
> + * the dcache flush. If the page is not uptodate, zero it to
> + * ensure we never go lacking for space here.
> + */
> + if (!PageUptodate(page)) {
> + void *kaddr = kmap_local_page(page);
> +
> + memset(kaddr, 0, PAGE_SIZE);
> + SetPageUptodate(page);
> + kunmap_local(kaddr);
> + }
Does the xfiles implementation prevent THPs from being created?
If not, this could lead to an entire THP being marked uptodate even
though we've only zeroed one page of it.
