--- name: "uiomove_frombuf" section: "9" platform: "freebsd" package: "base" version: "16.0-CURRENT-20261027-61c70dc16a79" release: "FreeBSD 16.0-CURRENT (2026-10-27)" platforms: ["freebsd"] url: "https://new.carta.tech/uio" --- # uiomove\_frombuf(9) device driver I/O routines Part of [uio(9)](/uio/9.md). ## NAME **uio**, **uiomove**, **uiomove\_frombuf**, **uiomove\_fromphys**, **uiomove\_nofault** – device driver I/O routines ## SYNOPSIS **#include \** **#include \** ``` struct uio { struct iovec *uio_iov; /* scatter/gather list */ int uio_iovcnt; /* length of scatter/gather list */ off_t uio_offset; /* offset in target object */ ssize_t uio_resid; /* remaining bytes to process */ enum uio_seg uio_segflg; /* address space */ enum uio_rw uio_rw; /* operation */ struct thread *uio_td; /* owner */ }; ``` *int*\ **uiomove**(*void \*buf*, *int howmuch*, *struct uio \*uiop*); *int*\ **uiomove\_frombuf**(*void \*buf*, *int howmuch*, *struct uio \*uiop*); *int*\ **uiomove\_fromphys**(*vm\_page\_t ma\[\]*, *vm\_offset\_t offset*, *int howmuch*, *struct uio \*uiop*); *int*\ **uiomove\_nofault**(*void \*buf*, *int howmuch*, *struct uio \*uiop*); ## DESCRIPTION The functions **uiomove**(), **uiomove\_frombuf**(), **uiomove\_fromphys**(), and **uiomove\_nofault**() are used to transfer data between buffers and I/O vectors that might possibly cross the user/kernel space boundary. As a result of any [read(2)](/read/2.md), [write(2)](/write/2.md), [readv(2)](/readv/2.md), or [writev(2)](/writev/2.md) system call that is being passed to a character-device driver, the appropriate driver *d\_read* or *d\_write* entry will be called with a pointer to a *struct uio* being passed. The transfer request is encoded in this structure. The driver itself should use **uiomove**() or **uiomove\_nofault**() to get at the data in this structure. The fields in the *uio* structure are: - *uio\_iov* The array of I/O vectors to be processed. In the case of scatter/gather I/O, this will be more than one vector. - *uio\_iovcnt* The number of I/O vectors present. - *uio\_offset* The offset into the device. - *uio\_resid* The remaining number of bytes to process, updated after transfer. - *uio\_segflg* One of the following flags: - UIO\_USERSPACE The I/O vector points into a process's address space. - UIO\_SYSSPACE The I/O vector points into the kernel address space. - UIO\_NOCOPY Do not copy, already in object. - *uio\_rw* The direction of the desired transfer. The supported flags are: - UIO\_READ Transfer data from the buffers into the I/O vectors. - UIO\_WRITE Transfer data from the I/O vectors into the buffers. - *uio\_td* The pointer to a *struct thread* for the associated thread; used if *uio\_segflg* indicates that the transfer is to be made from/to a process's address space. The function **uiomove\_nofault**() requires that the buffer and I/O vectors be accessible without incurring a page fault. The source and destination addresses must be physically mapped for read and write access, respectively, and neither the source nor destination addresses may be pageable. Thus, the function **uiomove\_nofault**() can be called from contexts where acquiring virtual memory system locks or sleeping are prohibited. The **uiomove\_frombuf**() function is a convenience wrapper around **uiomove**() for drivers that serve data which is wholly contained within an existing buffer in memory. It validates the *uio\_offset* and *uio\_resid* values against the size of the existing buffer, handling short transfers when the request partially overlaps the buffer. When *uio\_offset* is greater than or equal to the buffer size, the result is success with no bytes transferred, effectively signaling EOF. The **uiomove\_fromphys**() function provides a machine-independent way to copy memory to and from the physical address space. The *ma\[\]* argument is an array of physical pages. Every physical page address in the array provides a page-sized chunk of the physical space. The *offset* argument is the offset into the *ma\[\]* array. In particular, the offset does not have to lie within the first page. ## RETURN VALUES On success **uiomove**(), **uiomove\_frombuf**(), **uiomove\_fromphys**(), and **uiomove\_nofault**() will return 0; on error they will return an appropriate error code. ## EXAMPLES The idea is that the driver maintains a private buffer for its data, and processes the request in chunks of maximal the size of this buffer. Note that the buffer handling below is very simplified and will not work (the buffer pointer is not being advanced in case of a partial read), it is just here to demonstrate the **uio** handling. ``` /* MIN() can be found there: */ #include #define BUFSIZE 512 static char buffer[BUFSIZE]; static int data_available; /* amount of data that can be read */ static int fooread(struct cdev *dev, struct uio *uio, int flag) { int rv, amnt; rv = 0; while (uio->uio_resid > 0) { if (data_available > 0) { amnt = MIN(uio->uio_resid, data_available); rv = uiomove(buffer, amnt, uio); if (rv != 0) break; data_available -= amnt; } else tsleep(...); /* wait for a better time */ } if (rv != 0) { /* do error cleanup here */ } return (rv); } ``` ## ERRORS **uiomove**(), **uiomove\_fromphys**() and **uiomove\_nofault**() will fail and return the following error code if: - \[EFAULT\] The invoked [copyin(9)](/copyin/9.md) or [copyout(9)](/copyout/9.md) returned EFAULT. In addition, **uiomove\_nofault**() will fail and return the following error code if: - \[EFAULT\] A page fault occurs. ## SEE ALSO [read(2)](/read/2.md), [readv(2)](/readv/2.md), [write(2)](/write/2.md), [writev(2)](/writev/2.md), [copyin(9)](/copyin/9.md), [copyout(9)](/copyout/9.md), [sleep(9)](/sleep/9.md) ## HISTORY The **uio** mechanism appeared in some early version of UNIX. ## AUTHORS This manual page was written by Jörg Wunsch.