freebsd-dev/sys/kern/subr_uio.c

571 lines
12 KiB
C
Raw Normal View History

/*-
1994-05-24 10:09:53 +00:00
* Copyright (c) 1982, 1986, 1991, 1993
* The Regents of the University of California. All rights reserved.
* (c) UNIX System Laboratories, Inc.
* All or some portions of this file are derived from material licensed
* to the University of California by American Telephone and Telegraph
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
* the permission of UNIX System Laboratories, Inc.
*
* Copyright (c) 2014 The FreeBSD Foundation
*
* Portions of this software were developed by Konstantin Belousov
* under sponsorship from the FreeBSD Foundation.
*
1994-05-24 10:09:53 +00:00
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 4. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)kern_subr.c 8.3 (Berkeley) 1/21/94
*/
2003-06-11 00:56:59 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/limits.h>
#include <sys/lock.h>
#include <sys/mman.h>
1994-05-24 10:09:53 +00:00
#include <sys/proc.h>
#include <sys/resourcevar.h>
Switch the vm_object mutex to be a rwlock. This will enable in the future further optimizations where the vm_object lock will be held in read mode most of the time the page cache resident pool of pages are accessed for reading purposes. The change is mostly mechanical but few notes are reported: * The KPI changes as follow: - VM_OBJECT_LOCK() -> VM_OBJECT_WLOCK() - VM_OBJECT_TRYLOCK() -> VM_OBJECT_TRYWLOCK() - VM_OBJECT_UNLOCK() -> VM_OBJECT_WUNLOCK() - VM_OBJECT_LOCK_ASSERT(MA_OWNED) -> VM_OBJECT_ASSERT_WLOCKED() (in order to avoid visibility of implementation details) - The read-mode operations are added: VM_OBJECT_RLOCK(), VM_OBJECT_TRYRLOCK(), VM_OBJECT_RUNLOCK(), VM_OBJECT_ASSERT_RLOCKED(), VM_OBJECT_ASSERT_LOCKED() * The vm/vm_pager.h namespace pollution avoidance (forcing requiring sys/mutex.h in consumers directly to cater its inlining functions using VM_OBJECT_LOCK()) imposes that all the vm/vm_pager.h consumers now must include also sys/rwlock.h. * zfs requires a quite convoluted fix to include FreeBSD rwlocks into the compat layer because the name clash between FreeBSD and solaris versions must be avoided. At this purpose zfs redefines the vm_object locking functions directly, isolating the FreeBSD components in specific compat stubs. The KPI results heavilly broken by this commit. Thirdy part ports must be updated accordingly (I can think off-hand of VirtualBox, for example). Sponsored by: EMC / Isilon storage division Reviewed by: jeff Reviewed by: pjd (ZFS specific review) Discussed with: alc Tested by: pho
2013-03-09 02:32:23 +00:00
#include <sys/rwlock.h>
#include <sys/sched.h>
#include <sys/sysctl.h>
VM level code cleanups. 1) Start using TSM. Struct procs continue to point to upages structure, after being freed. Struct vmspace continues to point to pte object and kva space for kstack. u_map is now superfluous. 2) vm_map's don't need to be reference counted. They always exist either in the kernel or in a vmspace. The vmspaces are managed by reference counts. 3) Remove the "wired" vm_map nonsense. 4) No need to keep a cache of kernel stack kva's. 5) Get rid of strange looking ++var, and change to var++. 6) Change more data structures to use our "zone" allocator. Added struct proc, struct vmspace and struct vnode. This saves a significant amount of kva space and physical memory. Additionally, this enables TSM for the zone managed memory. 7) Keep ioopt disabled for now. 8) Remove the now bogus "single use" map concept. 9) Use generation counts or id's for data structures residing in TSM, where it allows us to avoid unneeded restart overhead during traversals, where blocking might occur. 10) Account better for memory deficits, so the pageout daemon will be able to make enough memory available (experimental.) 11) Fix some vnode locking problems. (From Tor, I think.) 12) Add a check in ufs_lookup, to avoid lots of unneeded calls to bcmp. (experimental.) 13) Significantly shrink, cleanup, and make slightly faster the vm_fault.c code. Use generation counts, get rid of unneded collpase operations, and clean up the cluster code. 14) Make vm_zone more suitable for TSM. This commit is partially as a result of discussions and contributions from other people, including DG, Tor Egge, PHK, and probably others that I have forgotten to attribute (so let me know, if I forgot.) This is not the infamous, final cleanup of the vnode stuff, but a necessary step. Vnode mgmt should be correct, but things might still change, and there is still some missing stuff (like ioopt, and physical backing of non-merged cache files, debugging of layering concepts.)
1998-01-22 17:30:44 +00:00
#include <sys/vnode.h>
#include <vm/vm.h>
#include <vm/vm_param.h>
#include <vm/vm_extern.h>
#include <vm/vm_page.h>
#include <vm/vm_pageout.h>
#include <vm/vm_map.h>
1994-05-24 10:09:53 +00:00
SYSCTL_INT(_kern, KERN_IOV_MAX, iov_max, CTLFLAG_RD, SYSCTL_NULL_INT_PTR, UIO_MAXIOV,
"Maximum number of elements in an I/O vector; sysconf(_SC_IOV_MAX)");
static int uiomove_faultflag(void *cp, int n, struct uio *uio, int nofault);
int
copyin_nofault(const void *udaddr, void *kaddr, size_t len)
{
int error, save;
save = vm_fault_disable_pagefaults();
error = copyin(udaddr, kaddr, len);
vm_fault_enable_pagefaults(save);
return (error);
}
int
copyout_nofault(const void *kaddr, void *udaddr, size_t len)
{
int error, save;
save = vm_fault_disable_pagefaults();
error = copyout(kaddr, udaddr, len);
vm_fault_enable_pagefaults(save);
return (error);
}
#define PHYS_PAGE_COUNT(len) (howmany(len, PAGE_SIZE) + 1)
int
physcopyin(void *src, vm_paddr_t dst, size_t len)
{
vm_page_t m[PHYS_PAGE_COUNT(len)];
struct iovec iov[1];
struct uio uio;
int i;
iov[0].iov_base = src;
iov[0].iov_len = len;
uio.uio_iov = iov;
uio.uio_iovcnt = 1;
uio.uio_offset = 0;
uio.uio_resid = len;
uio.uio_segflg = UIO_SYSSPACE;
uio.uio_rw = UIO_WRITE;
for (i = 0; i < PHYS_PAGE_COUNT(len); i++, dst += PAGE_SIZE)
m[i] = PHYS_TO_VM_PAGE(dst);
return (uiomove_fromphys(m, dst & PAGE_MASK, len, &uio));
}
int
physcopyout(vm_paddr_t src, void *dst, size_t len)
{
vm_page_t m[PHYS_PAGE_COUNT(len)];
struct iovec iov[1];
struct uio uio;
int i;
iov[0].iov_base = dst;
iov[0].iov_len = len;
uio.uio_iov = iov;
uio.uio_iovcnt = 1;
uio.uio_offset = 0;
uio.uio_resid = len;
uio.uio_segflg = UIO_SYSSPACE;
uio.uio_rw = UIO_READ;
for (i = 0; i < PHYS_PAGE_COUNT(len); i++, src += PAGE_SIZE)
m[i] = PHYS_TO_VM_PAGE(src);
return (uiomove_fromphys(m, src & PAGE_MASK, len, &uio));
}
#undef PHYS_PAGE_COUNT
int
uiomove(void *cp, int n, struct uio *uio)
1994-05-24 10:09:53 +00:00
{
return (uiomove_faultflag(cp, n, uio, 0));
}
int
uiomove_nofault(void *cp, int n, struct uio *uio)
{
return (uiomove_faultflag(cp, n, uio, 1));
}
static int
uiomove_faultflag(void *cp, int n, struct uio *uio, int nofault)
{
struct thread *td;
struct iovec *iov;
size_t cnt;
int error, newflags, save;
td = curthread;
error = 0;
1994-05-24 10:09:53 +00:00
KASSERT(uio->uio_rw == UIO_READ || uio->uio_rw == UIO_WRITE,
1999-01-10 01:58:29 +00:00
("uiomove: mode"));
KASSERT(uio->uio_segflg != UIO_USERSPACE || uio->uio_td == td,
1999-01-10 01:58:29 +00:00
("uiomove proc"));
if (!nofault)
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, NULL,
"Calling uiomove()");
/* XXX does it make a sense to set TDP_DEADLKTREAT for UIO_SYSSPACE ? */
newflags = TDP_DEADLKTREAT;
if (uio->uio_segflg == UIO_USERSPACE && nofault) {
/*
* Fail if a non-spurious page fault occurs.
*/
newflags |= TDP_NOFAULTING | TDP_RESETSPUR;
}
save = curthread_pflags_set(newflags);
1994-05-24 10:09:53 +00:00
while (n > 0 && uio->uio_resid) {
iov = uio->uio_iov;
cnt = iov->iov_len;
if (cnt == 0) {
uio->uio_iov++;
uio->uio_iovcnt--;
continue;
}
if (cnt > n)
cnt = n;
1995-02-12 09:11:47 +00:00
1994-05-24 10:09:53 +00:00
switch (uio->uio_segflg) {
case UIO_USERSPACE:
maybe_yield();
1994-05-24 10:09:53 +00:00
if (uio->uio_rw == UIO_READ)
error = copyout(cp, iov->iov_base, cnt);
else
error = copyin(iov->iov_base, cp, cnt);
if (error)
goto out;
1994-05-24 10:09:53 +00:00
break;
case UIO_SYSSPACE:
if (uio->uio_rw == UIO_READ)
bcopy(cp, iov->iov_base, cnt);
1994-05-24 10:09:53 +00:00
else
bcopy(iov->iov_base, cp, cnt);
break;
case UIO_NOCOPY:
break;
}
iov->iov_base = (char *)iov->iov_base + cnt;
iov->iov_len -= cnt;
uio->uio_resid -= cnt;
uio->uio_offset += cnt;
cp = (char *)cp + cnt;
n -= cnt;
}
out:
curthread_pflags_restore(save);
return (error);
}
/*
* Wrapper for uiomove() that validates the arguments against a known-good
* kernel buffer. Currently, uiomove accepts a signed (n) argument, which
* is almost definitely a bad thing, so we catch that here as well. We
* return a runtime failure, but it might be desirable to generate a runtime
* assertion failure instead.
*/
int
uiomove_frombuf(void *buf, int buflen, struct uio *uio)
{
size_t offset, n;
if (uio->uio_offset < 0 || uio->uio_resid < 0 ||
(offset = uio->uio_offset) != uio->uio_offset)
return (EINVAL);
if (buflen <= 0 || offset >= buflen)
return (0);
if ((n = buflen - offset) > IOSIZE_MAX)
return (EINVAL);
return (uiomove((char *)buf + offset, n, uio));
}
1994-05-24 10:09:53 +00:00
/*
* Give next character to user as result of read.
*/
int
ureadc(int c, struct uio *uio)
1994-05-24 10:09:53 +00:00
{
struct iovec *iov;
char *iov_base;
1994-05-24 10:09:53 +00:00
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, NULL,
"Calling ureadc()");
1994-05-24 10:09:53 +00:00
again:
if (uio->uio_iovcnt == 0 || uio->uio_resid == 0)
panic("ureadc");
iov = uio->uio_iov;
if (iov->iov_len == 0) {
uio->uio_iovcnt--;
uio->uio_iov++;
goto again;
}
switch (uio->uio_segflg) {
case UIO_USERSPACE:
if (subyte(iov->iov_base, c) < 0)
return (EFAULT);
break;
case UIO_SYSSPACE:
iov_base = iov->iov_base;
*iov_base = c;
1994-05-24 10:09:53 +00:00
break;
case UIO_NOCOPY:
break;
1994-05-24 10:09:53 +00:00
}
iov->iov_base = (char *)iov->iov_base + 1;
1994-05-24 10:09:53 +00:00
iov->iov_len--;
uio->uio_resid--;
uio->uio_offset++;
return (0);
}
int
copyinfrom(const void * __restrict src, void * __restrict dst, size_t len,
int seg)
{
int error = 0;
switch (seg) {
case UIO_USERSPACE:
error = copyin(src, dst, len);
break;
case UIO_SYSSPACE:
bcopy(src, dst, len);
break;
default:
panic("copyinfrom: bad seg %d\n", seg);
}
return (error);
}
int
copyinstrfrom(const void * __restrict src, void * __restrict dst, size_t len,
size_t * __restrict copied, int seg)
{
int error = 0;
switch (seg) {
case UIO_USERSPACE:
error = copyinstr(src, dst, len, copied);
break;
case UIO_SYSSPACE:
error = copystr(src, dst, len, copied);
break;
default:
panic("copyinstrfrom: bad seg %d\n", seg);
}
return (error);
}
int
copyiniov(const struct iovec *iovp, u_int iovcnt, struct iovec **iov, int error)
{
u_int iovlen;
*iov = NULL;
if (iovcnt > UIO_MAXIOV)
return (error);
iovlen = iovcnt * sizeof (struct iovec);
*iov = malloc(iovlen, M_IOV, M_WAITOK);
error = copyin(iovp, *iov, iovlen);
if (error) {
free(*iov, M_IOV);
*iov = NULL;
}
return (error);
}
int
copyinuio(const struct iovec *iovp, u_int iovcnt, struct uio **uiop)
{
2004-02-04 08:14:47 +00:00
struct iovec *iov;
struct uio *uio;
u_int iovlen;
2004-02-04 08:14:47 +00:00
int error, i;
*uiop = NULL;
if (iovcnt > UIO_MAXIOV)
return (EINVAL);
2004-02-04 08:14:47 +00:00
iovlen = iovcnt * sizeof (struct iovec);
uio = malloc(iovlen + sizeof *uio, M_IOV, M_WAITOK);
iov = (struct iovec *)(uio + 1);
error = copyin(iovp, iov, iovlen);
if (error) {
free(uio, M_IOV);
return (error);
2004-02-04 08:14:47 +00:00
}
uio->uio_iov = iov;
uio->uio_iovcnt = iovcnt;
uio->uio_segflg = UIO_USERSPACE;
uio->uio_offset = -1;
uio->uio_resid = 0;
for (i = 0; i < iovcnt; i++) {
if (iov->iov_len > IOSIZE_MAX - uio->uio_resid) {
free(uio, M_IOV);
return (EINVAL);
2004-02-04 08:14:47 +00:00
}
uio->uio_resid += iov->iov_len;
iov++;
}
*uiop = uio;
return (0);
}
struct uio *
cloneuio(struct uio *uiop)
{
struct uio *uio;
int iovlen;
iovlen = uiop->uio_iovcnt * sizeof (struct iovec);
uio = malloc(iovlen + sizeof *uio, M_IOV, M_WAITOK);
*uio = *uiop;
uio->uio_iov = (struct iovec *)(uio + 1);
bcopy(uiop->uio_iov, uio->uio_iov, iovlen);
return (uio);
}
/*
* Map some anonymous memory in user space of size sz, rounded up to the page
* boundary.
*/
int
copyout_map(struct thread *td, vm_offset_t *addr, size_t sz)
{
struct vmspace *vms;
int error;
vm_size_t size;
vms = td->td_proc->p_vmspace;
/*
* Map somewhere after heap in process memory.
*/
PROC_LOCK(td->td_proc);
*addr = round_page((vm_offset_t)vms->vm_daddr +
lim_max(td->td_proc, RLIMIT_DATA));
PROC_UNLOCK(td->td_proc);
/* round size up to page boundry */
size = (vm_size_t)round_page(sz);
error = vm_mmap(&vms->vm_map, addr, size, PROT_READ | PROT_WRITE,
VM_PROT_ALL, MAP_PRIVATE | MAP_ANON, OBJT_DEFAULT, NULL, 0);
return (error);
}
/*
* Unmap memory in user space.
*/
int
copyout_unmap(struct thread *td, vm_offset_t addr, size_t sz)
{
vm_map_t map;
vm_size_t size;
if (sz == 0)
return (0);
map = &td->td_proc->p_vmspace->vm_map;
size = (vm_size_t)round_page(sz);
if (vm_map_remove(map, addr, addr + size) != KERN_SUCCESS)
return (EINVAL);
return (0);
}
#ifdef NO_FUEWORD
/*
* XXXKIB The temporal implementation of fue*() functions which do not
* handle usermode -1 properly, mixing it with the fault code. Keep
* this until MD code is written. Currently sparc64, mips and arm do
* not have proper implementation.
*/
int
fueword(volatile const void *base, long *val)
{
long res;
res = fuword(base);
if (res == -1)
return (-1);
*val = res;
return (0);
}
int
fueword32(volatile const void *base, int32_t *val)
{
int32_t res;
res = fuword32(base);
if (res == -1)
return (-1);
*val = res;
return (0);
}
#ifdef _LP64
int
fueword64(volatile const void *base, int64_t *val)
{
int32_t res;
res = fuword64(base);
if (res == -1)
return (-1);
*val = res;
return (0);
}
#endif
int
casueword32(volatile uint32_t *base, uint32_t oldval, uint32_t *oldvalp,
uint32_t newval)
{
int32_t ov;
ov = casuword32(base, oldval, newval);
if (ov == -1)
return (-1);
*oldvalp = ov;
return (0);
}
int
casueword(volatile u_long *p, u_long oldval, u_long *oldvalp, u_long newval)
{
u_long ov;
ov = casuword(p, oldval, newval);
if (ov == -1)
return (-1);
*oldvalp = ov;
return (0);
}
#else /* NO_FUEWORD */
int32_t
fuword32(volatile const void *addr)
{
int rv;
int32_t val;
rv = fueword32(addr, &val);
return (rv == -1 ? -1 : val);
}
#ifdef _LP64
int64_t
fuword64(volatile const void *addr)
{
int rv;
int64_t val;
rv = fueword64(addr, &val);
return (rv == -1 ? -1 : val);
}
#endif /* _LP64 */
long
fuword(volatile const void *addr)
{
long val;
int rv;
rv = fueword(addr, &val);
return (rv == -1 ? -1 : val);
}
uint32_t
casuword32(volatile uint32_t *addr, uint32_t old, uint32_t new)
{
int rv;
uint32_t val;
rv = casueword32(addr, old, &val, new);
return (rv == -1 ? -1 : val);
}
u_long
casuword(volatile u_long *addr, u_long old, u_long new)
{
int rv;
u_long val;
rv = casueword(addr, old, &val, new);
return (rv == -1 ? -1 : val);
}
#endif /* NO_FUEWORD */