File:  [DragonFly] / src / sys / kern / kern_physio.c
Revision 1.5: download - view: text, annotated - select for diffs
Tue Jul 22 17:03:33 2003 UTC (11 years, 11 months ago) by dillon
Branches: MAIN
CVS tags: HEAD
DEV messaging stage 2/4: In this stage all DEV commands are now being
funneled through the message port for action by the port's beginmsg function.
CONSOLE and DISK device shims replace the port with their own and then
forward to the original.  FB (Frame Buffer) shims supposedly do the same
thing but I haven't been able to test it.   I don't expect instability
in mainline code but there might be easy-to-fix, and some drivers still need
to be converted.  See primarily: kern/kern_device.c (new dev_*() functions and
inherits cdevsw code from kern/kern_conf.c), sys/device.h, and kern/subr_disk.c
for the high points.

In this stage all DEV messages are still acted upon synchronously in the
context of the caller.  We cannot create a separate handler thread until
the copyin's (primarily in ioctl functions) are made thread-aware.

Note that the messaging shims are going to look rather messy in these early
days but as more subsystems are converted over we will begin to use
pre-initialized messages and message forwarding to avoid having to constantly
rebuild messages prior to use.

Note that DEV itself is a mess oweing to its 4.x roots and will be cleaned
up in subsequent passes.  e.g. the way sub-devices inherit the main device's
cdevsw was always a bad hack and it still is, and several functions
(mmap, kqfilter, psize, poll) return results rather then error codes, which
will be fixed since now we have a message to store the result in :-)

 * Copyright (c) 1994 John S. Dyson
 * All rights reserved.
 * 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 immediately at the beginning of the file, without modification,
 *    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.
 * 3. Absolutely no warranty of function or purpose is made by the author
 *    John S. Dyson.
 * 4. Modifications may be freely made to this file if the above conditions
 *    are met.
 * $FreeBSD: src/sys/kern/kern_physio.c,v 2003/05/29 06:15:35 alc Exp $
 * $DragonFly: src/sys/kern/kern_physio.c,v 1.5 2003/07/22 17:03:33 dillon Exp $

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/buf.h>
#include <sys/conf.h>
#include <sys/proc.h>
#include <sys/uio.h>
#include <sys/device.h>

#include <vm/vm.h>
#include <vm/vm_extern.h>

static void
physwakeup(struct buf *bp)
	wakeup((caddr_t) bp);

physio(dev_t dev, struct uio *uio, int ioflag)
	int i;
	int error;
	int spl;
	caddr_t sa;
	off_t blockno;
	u_int iolen;
	struct buf *bp;

	 * NOTE: we no longer have to PHOLD() the process, because the
	 * kernel stack / uarea cannot be swapped, and when it can be in
	 * the future it will only happen if the process is sleeping on
	 * a particular address.
	bp = getpbuf(NULL);
	sa = bp->b_data;
	error = 0;

	/* XXX: sanity check */
	if(dev->si_iosize_max < PAGE_SIZE) {
		printf("WARNING: %s si_iosize_max=%d, using DFLTPHYS.\n",
		    devtoname(dev), dev->si_iosize_max);
		dev->si_iosize_max = DFLTPHYS;

	for (i = 0; i < uio->uio_iovcnt; i++) {
		while (uio->uio_iov[i].iov_len) {
			if (uio->uio_rw == UIO_READ)
				bp->b_flags = B_PHYS | B_CALL | B_READ;
				bp->b_flags = B_PHYS | B_CALL | B_WRITE;
			bp->b_dev = dev;
			bp->b_iodone = physwakeup;
			bp->b_data = uio->uio_iov[i].iov_base;
			bp->b_bcount = uio->uio_iov[i].iov_len;
			bp->b_offset = uio->uio_offset;
			bp->b_saveaddr = sa;

			/* Don't exceed drivers iosize limit */
			if (bp->b_bcount > dev->si_iosize_max)
				bp->b_bcount = dev->si_iosize_max;

			 * Make sure the pbuf can map the request
			 * XXX: The pbuf has kvasize = MAXPHYS so a request
			 * XXX: larger than MAXPHYS - PAGE_SIZE must be
			 * XXX: page aligned or it will be fragmented.
			iolen = ((vm_offset_t) bp->b_data) & PAGE_MASK;
			if ((bp->b_bcount + iolen) > bp->b_kvasize) {
				bp->b_bcount = bp->b_kvasize;
				if (iolen != 0)
					bp->b_bcount -= PAGE_SIZE;
			bp->b_bufsize = bp->b_bcount;

			blockno = bp->b_offset >> DEV_BSHIFT;
			if ((daddr_t)blockno != blockno) {
				error = EINVAL; /* blockno overflow */
				goto doerror;
			bp->b_blkno = blockno;

			if (uio->uio_segflg == UIO_USERSPACE)
				if (vmapbuf(bp) < 0) {
					error = EFAULT;
					goto doerror;

			BUF_STRATEGY(bp, 0);
			spl = splbio();
			while ((bp->b_flags & B_DONE) == 0)
				tsleep((caddr_t)bp, 0, "physstr", 0);

			if (uio->uio_segflg == UIO_USERSPACE)
			iolen = bp->b_bcount - bp->b_resid;
			if (iolen == 0 && !(bp->b_flags & B_ERROR))
				goto doerror;	/* EOF */
			uio->uio_iov[i].iov_len -= iolen;
			uio->uio_iov[i].iov_base += iolen;
			uio->uio_resid -= iolen;
			uio->uio_offset += iolen;
			if( bp->b_flags & B_ERROR) {
				error = bp->b_error;
				goto doerror;
	relpbuf(bp, NULL);
	return (error);