diff --git a/mm/filemap.c b/mm/filemap.c index 529eb9e..5f189e0 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1170,6 +1170,31 @@ success: return size; } +int generic_segment_checks(const struct iovec *iov, unsigned long *nr_segs, + size_t *count, unsigned long access_flags) +{ + unsigned long seg; + for (seg = 0; seg < *nr_segs; seg++) { + const struct iovec *iv = &iov[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + *count += iv->iov_len; + if (unlikely((ssize_t)(*count|iv->iov_len) < 0)) + return -EINVAL; + if (access_ok(access_flags, iv->iov_base, iv->iov_len)) + continue; + if (seg == 0) + return -EFAULT; + *nr_segs = seg; + *count -= iv->iov_len; /* This segment is no good */ + break; + } + return 0; +} + /** * generic_file_aio_read - generic filesystem read routine * @iocb: kernel I/O control block @@ -1191,24 +1216,9 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, loff_t *ppos = &iocb->ki_pos; count = 0; - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *iv = &iov[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - count += iv->iov_len; - if (unlikely((ssize_t)(count|iv->iov_len) < 0)) - return -EINVAL; - if (access_ok(VERIFY_WRITE, iv->iov_base, iv->iov_len)) - continue; - if (seg == 0) - return -EFAULT; - nr_segs = seg; - count -= iv->iov_len; /* This segment is no good */ - break; - } + retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); + if (retval) + return retval; /* coalesce the iovecs and go direct-to-BIO for O_DIRECT */ if (filp->f_flags & O_DIRECT) { @@ -2103,8 +2113,9 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, /* * Sync the fs metadata but not the minor inode changes and * of course not the data as we did direct DMA for the IO. - * i_mutex is held, which protects generic_osync_inode() from - * livelocking. AIO O_DIRECT ops attempt to sync metadata here. + * i_mutex may not being held, if so some specific locking + * ordering must protect generic_osync_inode() from livelocking. + * AIO O_DIRECT ops attempt to sync metadata here. */ if ((written >= 0 || written == -EIOCBQUEUED) && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { @@ -2293,30 +2304,14 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov, size_t ocount; /* original count */ size_t count; /* after file limit checks */ struct inode *inode = mapping->host; - unsigned long seg; loff_t pos; ssize_t written; ssize_t err; ocount = 0; - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *iv = &iov[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - ocount += iv->iov_len; - if (unlikely((ssize_t)(ocount|iv->iov_len) < 0)) - return -EINVAL; - if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) - continue; - if (seg == 0) - return -EFAULT; - nr_segs = seg; - ocount -= iv->iov_len; /* This segment is no good */ - break; - } + err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ); + if (err) + return err; if (unlikely(aio_restarted())) { /* nothing to transfer, may just need to sync data */ return ocount; @@ -2442,6 +2437,20 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, mutex_lock(&inode->i_mutex); ret = __generic_file_aio_write_nolock(iocb, iov, nr_segs, &iocb->ki_pos); + if (unlikely( ret < 0 && file->f_flags & O_DIRECT)) { + unsigned long nr_segs_avail = nr_segs; + size_t count = 0; + if (!generic_segment_checks(iov, &nr_segs_avail, &count, + VERIFY_READ)) { + loff_t isize = i_size_read(inode); + /* + * direct_IO() may have instantiated a few blocks + * outside i_size. Trim these off again. + */ + if (pos + count > isize) + vmtruncate(inode, isize); + } + } mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { @@ -2458,8 +2467,8 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, EXPORT_SYMBOL(generic_file_aio_write); /* - * Called under i_mutex for writes to S_ISREG files. Returns -EIO if something - * went wrong during pagecache shootdown. + * May be called without i_mutex for writes to S_ISREG files. + * Returns -EIO if something went wrong during pagecache shootdown. */ static ssize_t generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,