1.\" Copyright (c) 1998 2.\" The Regents of the University of California. All rights reserved. 3.\" 4.\" Redistribution and use in source and binary forms, with or without 5.\" modification, are permitted provided that the following conditions 6.\" are met: 7.\" 1. Redistributions of source code must retain the above copyright 8.\" notice, this list of conditions and the following disclaimer. 9.\" 2. Redistributions in binary form must reproduce the above copyright 10.\" notice, this list of conditions and the following disclaimer in the 11.\" documentation and/or other materials provided with the distribution. 12.\" 3. All advertising materials mentioning features or use of this software 13.\" must display the following acknowledgement: 14.\" This product includes software developed by the University of 15.\" California, Berkeley and its contributors. 16.\" 4. Neither the name of the University nor the names of its contributors 17.\" may be used to endorse or promote products derived from this software 18.\" without specific prior written permission. 19.\" 20.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 23.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 30.\" SUCH DAMAGE. 31.\" 32.\" $FreeBSD$ 33.\" 34.Dd December 22, 1998 35.Dt BUF 9 36.Os 37.Sh NAME 38.Nm buf 39.Nd "kernel buffer I/O scheme used in FreeBSD VM system" 40.Sh DESCRIPTION 41The kernel implements a KVM abstraction of the buffer cache which allows it 42to map potentially disparate vm_page's into contiguous KVM for use by 43(mainly file system) devices and device I/O. 44This abstraction supports 45block sizes from DEV_BSIZE (usually 512) to upwards of several pages or more. 46It also supports a relatively primitive byte-granular valid range and dirty 47range currently hardcoded for use by NFS. 48The code implementing the 49VM Buffer abstraction is mostly concentrated in 50.Pa /usr/src/sys/kern/vfs_bio.c . 51.Pp 52One of the most important things to remember when dealing with buffer pointers 53(struct buf) is that the underlying pages are mapped directly from the buffer 54cache. 55No data copying occurs in the scheme proper, though some file systems 56such as UFS do have to copy a little when dealing with file fragments. 57The second most important thing to remember is that due to the underlying page 58mapping, the b_data base pointer in a buf is always *page* aligned, not 59*block* aligned. 60When you have a VM buffer representing some b_offset and 61b_size, the actual start of the buffer is (b_data + (b_offset & PAGE_MASK)) 62and not just b_data. 63Finally, the VM system's core buffer cache supports 64valid and dirty bits (m->valid, m->dirty) for pages in DEV_BSIZE chunks. 65Thus 66a platform with a hardware page size of 4096 bytes has 8 valid and 8 dirty 67bits. 68These bits are generally set and cleared in groups based on the device 69block size of the device backing the page. 70Complete page's worth are often 71referred to using the VM_PAGE_BITS_ALL bitmask (i.e., 0xFF if the hardware page 72size is 4096). 73.Pp 74VM buffers also keep track of a byte-granular dirty range and valid range. 75This feature is normally only used by the NFS subsystem. 76I'm not sure why it 77is used at all, actually, since we have DEV_BSIZE valid/dirty granularity 78within the VM buffer. 79If a buffer dirty operation creates a 'hole', 80the dirty range will extend to cover the hole. 81If a buffer validation 82operation creates a 'hole' the byte-granular valid range is left alone and 83will not take into account the new extension. 84Thus the whole byte-granular 85abstraction is considered a bad hack and it would be nice if we could get rid 86of it completely. 87.Pp 88A VM buffer is capable of mapping the underlying VM cache pages into KVM in 89order to allow the kernel to directly manipulate the data associated with 90the (vnode,b_offset,b_size). 91The kernel typically unmaps VM buffers the moment 92they are no longer needed but often keeps the 'struct buf' structure 93instantiated and even bp->b_pages array instantiated despite having unmapped 94them from KVM. 95If a page making up a VM buffer is about to undergo I/O, the 96system typically unmaps it from KVM and replaces the page in the b_pages[] 97array with a place-marker called bogus_page. 98The place-marker forces any kernel 99subsystems referencing the associated struct buf to re-lookup the associated 100page. 101I believe the place-marker hack is used to allow sophisticated devices 102such as file system devices to remap underlying pages in order to deal with, 103for example, re-mapping a file fragment into a file block. 104.Pp 105VM buffers are used to track I/O operations within the kernel. 106Unfortunately, 107the I/O implementation is also somewhat of a hack because the kernel wants 108to clear the dirty bit on the underlying pages the moment it queues the I/O 109to the VFS device, not when the physical I/O is actually initiated. 110This 111can create confusion within file system devices that use delayed-writes because 112you wind up with pages marked clean that are actually still dirty. 113If not 114treated carefully, these pages could be thrown away! 115Indeed, a number of 116serious bugs related to this hack were not fixed until the 2.2.8/3.0 release. 117The kernel uses an instantiated VM buffer (i.e., struct buf) to place-mark pages 118in this special state. 119The buffer is typically flagged B_DELWRI. 120When a 121device no longer needs a buffer it typically flags it as B_RELBUF. 122Due to 123the underlying pages being marked clean, the B_DELWRI|B_RELBUF combination must 124be interpreted to mean that the buffer is still actually dirty and must be 125written to its backing store before it can actually be released. 126In the case 127where B_DELWRI is not set, the underlying dirty pages are still properly 128marked as dirty and the buffer can be completely freed without losing that 129clean/dirty state information. 130(XXX do we have to check other flags in 131regards to this situation ???) 132.Pp 133The kernel reserves a portion of its KVM space to hold VM Buffer's data 134maps. 135Even though this is virtual space (since the buffers are mapped 136from the buffer cache), we cannot make it arbitrarily large because 137instantiated VM Buffers (struct buf's) prevent their underlying pages in the 138buffer cache from being freed. 139This can complicate the life of the paging 140system. 141.Pp 142.\" .Sh SEE ALSO 143.\" .Xr <fillmein> 9 144.Sh HISTORY 145The 146.Nm 147manual page was originally written by 148.An Matthew Dillon 149and first appeared in 150.Fx 3.1 , 151December 1998. 152