minix/servers/vfs/read.c

344 lines
8.7 KiB
C
Raw Normal View History

/* This file contains the heart of the mechanism used to read (and write)
* files. Read and write requests are split up into chunks that do not cross
* block boundaries. Each chunk is then processed in turn. Reads on special
* files are also detected and handled.
*
* The entry points into this file are
* do_read: perform the READ system call by calling read_write
* do_getdents: read entries from a directory (GETDENTS)
* read_write: actually do the work of READ and WRITE
*
*/
#include "fs.h"
#include <fcntl.h>
#include <unistd.h>
#include <minix/com.h>
#include <minix/u64.h>
#include "file.h"
#include "fproc.h"
2012-02-13 16:28:04 +01:00
#include "scratchpad.h"
#include "param.h"
#include <dirent.h>
Mostly bugfixes of bugs triggered by the test set. bugfixes: SYSTEM: . removed rc->p_priv->s_flags = 0; for the priv struct shared by all user processes in get_priv(). this should only be done once. doing a SYS_PRIV_USER in sys_privctl() caused the flags of all user processes to be reset, so they were no longer PREEMPTIBLE. this happened when RS executed a policy script. (this broke test1 in the test set) VFS/MFS: . chown can change the mode of a file, and chmod arguments are only part of the full file mode so the full filemode is slightly magic. changed these calls so that the final modes are returned to VFS, so that the vnode can be kept up-to-date. (this broke test11 in the test set) MFS: . lookup() checked for sizeof(string) instead of sizeof(user_path), truncating long path names (caught by test 23) . truncate functions neglected to update ctime (this broke test16) VFS: . corner case of an empty filename lookup caused fields of a request not to be filled in in the lookup functions, not making it clear that the lookup had failed, causing messages to garbage processes, causing strange failures. (caught by test 30) . trust v_size in vnode when doing reads or writes on non-special files, truncating i/o where necessary; this is necessary for pipes, as MFS can't tell when a pipe has been truncated without it being told explicitly each time. when the last reader/writer on a pipe closes, tell FS about the new size using truncate_vn(). (this broke test 25, among others) . permission check for chdir() had disappeared; added a forbidden() call (caught by test 23) new code, shouldn't change anything: . introduced RTS_SET, RTS_UNSET, and RTS_ISSET macro's, and their LOCK variants. These macros set and clear the p_rts_flags field, causing a lot of duplicated logic like old_flags = rp->p_rts_flags; /* save value of the flags */ rp->p_rts_flags &= ~NO_PRIV; if (old_flags != 0 && rp->p_rts_flags == 0) lock_enqueue(rp); to change into the simpler RTS_LOCK_UNSET(rp, NO_PRIV); so the macros take care of calling dequeue() and enqueue() (or lock_*()), as the case may be). This makes the code a bit more readable and a bit less fragile. . removed return code from do_clocktick in CLOCK as it currently never replies . removed some debug code from VFS . fixed grant debug message in device.c preemptive checks, tests, changes: . added return code checks of receive() to SYSTEM and CLOCK . O_TRUNC should never arrive at MFS (added sanity check and removed O_TRUNC code) . user_path declared with PATH_MAX+1 to let it be null-terminated . checks in MFS to see if strings passed by VFS are null-terminated IS: . static irq name table thrown out
2007-02-01 18:50:02 +01:00
#include <assert.h>
#include <minix/vfsif.h>
#include "vnode.h"
#include "vmnt.h"
/*===========================================================================*
* do_read *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
int do_read()
{
2012-02-13 16:28:04 +01:00
return(do_read_write(READING));
}
/*===========================================================================*
2012-02-13 16:28:04 +01:00
* lock_bsf *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
void lock_bsf(void)
2012-02-13 16:28:04 +01:00
{
struct fproc *org_fp;
struct worker_thread *org_self;
if (mutex_trylock(&bsf_lock) == 0)
return;
org_fp = fp;
org_self = self;
if (mutex_lock(&bsf_lock) != 0)
panic("unable to lock block special file lock");
fp = org_fp;
self = org_self;
}
/*===========================================================================*
* unlock_bsf *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
void unlock_bsf(void)
2012-02-13 16:28:04 +01:00
{
if (mutex_unlock(&bsf_lock) != 0)
panic("failed to unlock block special file lock");
}
/*===========================================================================*
* do_read_write *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
int do_read_write(rw_flag)
int rw_flag; /* READING or WRITING */
{
/* Perform read(fd, buffer, nbytes) or write(fd, buffer, nbytes) call. */
2012-02-13 16:28:04 +01:00
struct filp *f;
tll_access_t locktype;
int r;
scratch(fp).file.fd_nr = job_m_in.fd;
scratch(fp).io.io_buffer = job_m_in.buffer;
scratch(fp).io.io_nbytes = (size_t) job_m_in.nbytes;
2012-02-13 16:28:04 +01:00
locktype = (rw_flag == READING) ? VNODE_READ : VNODE_WRITE;
if ((f = get_filp(scratch(fp).file.fd_nr, locktype)) == NULL)
return(err_code);
if (((f->filp_mode) & (rw_flag == READING ? R_BIT : W_BIT)) == 0) {
2012-02-13 16:28:04 +01:00
unlock_filp(f);
return(f->filp_mode == FILP_CLOSED ? EIO : EBADF);
}
2012-02-13 16:28:04 +01:00
if (scratch(fp).io.io_nbytes == 0) {
unlock_filp(f);
return(0); /* so char special files need not check for 0*/
2012-02-13 16:28:04 +01:00
}
r = read_write(rw_flag, f, scratch(fp).io.io_buffer, scratch(fp).io.io_nbytes,
who_e);
unlock_filp(f);
return(r);
}
/*===========================================================================*
* read_write *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
int read_write(int rw_flag, struct filp *f, char *buf, size_t size,
2012-02-13 16:28:04 +01:00
endpoint_t for_e)
{
register struct vnode *vp;
u64_t position, res_pos, new_pos;
unsigned int cum_io, cum_io_incr, res_cum_io;
2012-04-25 14:44:42 +02:00
int op, oflags, r;
position = f->filp_pos;
oflags = f->filp_flags;
vp = f->filp_vno;
r = OK;
cum_io = 0;
2012-02-13 16:28:04 +01:00
if (size > SSIZE_MAX) return(EINVAL);
if (S_ISFIFO(vp->v_mode)) {
if (fp->fp_cum_io_partial != 0) {
2012-02-13 16:28:04 +01:00
panic("VFS: read_write: fp_cum_io_partial not clear");
2007-08-07 14:52:47 +02:00
}
2012-02-13 16:28:04 +01:00
r = rw_pipe(rw_flag, for_e, f, buf, size);
return(r);
}
op = (rw_flag == READING ? VFS_DEV_READ : VFS_DEV_WRITE);
2012-04-25 14:44:42 +02:00
if (S_ISCHR(vp->v_mode)) { /* Character special files. */
2008-02-22 15:26:41 +01:00
dev_t dev;
int suspend_reopen;
2012-04-25 14:44:42 +02:00
if (vp->v_sdev == NO_DEV)
panic("VFS: read_write tries to access char dev NO_DEV");
suspend_reopen = (f->filp_state != FS_NORMAL);
2008-02-22 15:26:41 +01:00
dev = (dev_t) vp->v_sdev;
2012-02-13 16:28:04 +01:00
r = dev_io(op, dev, for_e, buf, position, size, oflags,
suspend_reopen);
2008-02-22 15:26:41 +01:00
if (r >= 0) {
cum_io = r;
position = add64ul(position, r);
r = OK;
}
2012-04-25 14:44:42 +02:00
} else if (S_ISBLK(vp->v_mode)) { /* Block special files. */
if (vp->v_sdev == NO_DEV)
panic("VFS: read_write tries to access block dev NO_DEV");
2012-02-13 16:28:04 +01:00
lock_bsf();
r = req_breadwrite(vp->v_bfs_e, for_e, vp->v_sdev, position, size,
buf, rw_flag, &res_pos, &res_cum_io);
if (r == OK) {
position = res_pos;
cum_io += res_cum_io;
}
2012-02-13 16:28:04 +01:00
unlock_bsf();
} else { /* Regular files */
2012-04-25 14:44:42 +02:00
if (rw_flag == WRITING) {
/* Check for O_APPEND flag. */
if (oflags & O_APPEND) position = cvul64(vp->v_size);
}
2007-01-05 17:36:55 +01:00
/* Issue request */
2012-02-13 16:28:04 +01:00
r = req_readwrite(vp->v_fs_e, vp->v_inode_nr, position, rw_flag, for_e,
buf, size, &new_pos, &cum_io_incr);
if (r >= 0) {
if (ex64hi(new_pos))
panic("read_write: bad new pos");
position = new_pos;
cum_io += cum_io_incr;
}
}
/* On write, update file size and access time. */
if (rw_flag == WRITING) {
2012-04-25 14:44:42 +02:00
if (S_ISREG(vp->v_mode) || S_ISDIR(vp->v_mode)) {
if (cmp64ul(position, vp->v_size) > 0) {
if (ex64hi(position) != 0) {
panic("read_write: file size too big ");
}
vp->v_size = ex64lo(position);
}
}
}
f->filp_pos = position;
2012-02-13 16:28:04 +01:00
if (r == OK) return(cum_io);
return(r);
}
/*===========================================================================*
* do_getdents *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
int do_getdents()
{
/* Perform the getdents(fd, buf, size) system call. */
2012-02-13 16:28:04 +01:00
int r = OK;
u64_t new_pos;
register struct filp *rfilp;
scratch(fp).file.fd_nr = job_m_in.fd;
scratch(fp).io.io_buffer = job_m_in.buffer;
scratch(fp).io.io_nbytes = (size_t) job_m_in.nbytes;
/* Is the file descriptor valid? */
if ( (rfilp = get_filp(scratch(fp).file.fd_nr, VNODE_READ)) == NULL)
return(err_code);
2012-02-13 16:28:04 +01:00
if (!(rfilp->filp_mode & R_BIT))
2012-02-13 16:28:04 +01:00
r = EBADF;
2012-04-25 14:44:42 +02:00
else if (!S_ISDIR(rfilp->filp_vno->v_mode))
2012-02-13 16:28:04 +01:00
r = EBADF;
2012-02-13 16:28:04 +01:00
if (r == OK) {
if (ex64hi(rfilp->filp_pos) != 0)
panic("do_getdents: can't handle large offsets");
r = req_getdents(rfilp->filp_vno->v_fs_e, rfilp->filp_vno->v_inode_nr,
rfilp->filp_pos, scratch(fp).io.io_buffer,
scratch(fp).io.io_nbytes, &new_pos,0);
2012-02-13 16:28:04 +01:00
if (r > 0) rfilp->filp_pos = new_pos;
}
2012-02-13 16:28:04 +01:00
unlock_filp(rfilp);
return(r);
}
2007-08-07 14:52:47 +02:00
/*===========================================================================*
* rw_pipe *
*===========================================================================*/
2012-03-25 20:25:53 +02:00
int rw_pipe(rw_flag, usr_e, f, buf, req_size)
2007-08-07 14:52:47 +02:00
int rw_flag; /* READING or WRITING */
endpoint_t usr_e;
2007-08-07 14:52:47 +02:00
struct filp *f;
char *buf;
size_t req_size;
{
int r, oflags, partial_pipe = 0;
size_t size, cum_io, cum_io_incr;
2007-08-07 14:52:47 +02:00
struct vnode *vp;
u64_t position, new_pos;
2007-08-07 14:52:47 +02:00
2012-02-13 16:28:04 +01:00
/* Must make sure we're operating on locked filp and vnode */
assert(tll_islocked(&f->filp_vno->v_lock));
assert(mutex_trylock(&f->filp_lock) == -EDEADLK);
2007-08-07 14:52:47 +02:00
oflags = f->filp_flags;
vp = f->filp_vno;
position = cvu64((rw_flag == READING) ? vp->v_pipe_rd_pos :
2012-02-13 16:28:04 +01:00
vp->v_pipe_wr_pos);
2007-08-07 14:52:47 +02:00
/* fp->fp_cum_io_partial is only nonzero when doing partial writes */
2012-02-13 16:28:04 +01:00
cum_io = fp->fp_cum_io_partial;
2007-08-07 14:52:47 +02:00
r = pipe_check(vp, rw_flag, oflags, req_size, position, 0);
if (r <= 0) {
2012-02-13 16:28:04 +01:00
if (r == SUSPEND) pipe_suspend(f, buf, req_size);
return(r);
2007-08-07 14:52:47 +02:00
}
size = r;
if (size < req_size) partial_pipe = 1;
2007-08-07 14:52:47 +02:00
/* Truncate read request at size. */
if((rw_flag == READING) &&
2012-02-13 16:28:04 +01:00
cmp64ul(add64ul(position, size), vp->v_size) > 0) {
2007-08-07 14:52:47 +02:00
/* Position always should fit in an off_t (LONG_MAX). */
off_t pos32;
assert(cmp64ul(position, LONG_MAX) <= 0);
pos32 = cv64ul(position);
assert(pos32 >= 0);
assert(pos32 <= LONG_MAX);
size = vp->v_size - pos32;
}
2012-02-13 16:28:04 +01:00
if (vp->v_mapfs_e == 0)
panic("unmapped pipe");
2010-01-27 10:30:39 +01:00
r = req_readwrite(vp->v_mapfs_e, vp->v_mapinode_nr, position, rw_flag, usr_e,
buf, size, &new_pos, &cum_io_incr);
if (r >= 0) {
2007-08-07 14:52:47 +02:00
if (ex64hi(new_pos))
panic("rw_pipe: bad new pos");
2007-08-07 14:52:47 +02:00
position = new_pos;
cum_io += cum_io_incr;
buf += cum_io_incr;
req_size -= cum_io_incr;
}
2012-02-13 16:28:04 +01:00
2007-08-07 14:52:47 +02:00
/* On write, update file size and access time. */
if (rw_flag == WRITING) {
if (cmp64ul(position, vp->v_size) > 0) {
if (ex64hi(position) != 0) {
panic("read_write: file size too big for v_size");
2007-08-07 14:52:47 +02:00
}
vp->v_size = ex64lo(position);
}
} else {
2007-08-07 14:52:47 +02:00
if (cmp64ul(position, vp->v_size) >= 0) {
/* Reset pipe pointers */
vp->v_size = 0;
vp->v_pipe_rd_pos= 0;
vp->v_pipe_wr_pos= 0;
2007-08-07 14:52:47 +02:00
position = cvu64(0);
}
}
if (rw_flag == READING)
vp->v_pipe_rd_pos= cv64ul(position);
else
vp->v_pipe_wr_pos= cv64ul(position);
2007-08-07 14:52:47 +02:00
if (r == OK) {
if (partial_pipe) {
/* partial write on pipe with */
/* O_NONBLOCK, return write count */
if (!(oflags & O_NONBLOCK)) {
/* partial write on pipe with req_size > PIPE_SIZE,
* non-atomic
*/
fp->fp_cum_io_partial = cum_io;
2012-02-13 16:28:04 +01:00
pipe_suspend(f, buf, req_size);
2007-08-07 14:52:47 +02:00
return(SUSPEND);
}
}
fp->fp_cum_io_partial = 0;
return(cum_io);
2007-08-07 14:52:47 +02:00
}
return(r);
2007-08-07 14:52:47 +02:00
}