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 41.Pp 42The kernel implements a KVM abstraction of the buffer cache which allows it 43to map potentially disparate vm_page's into contiguous KVM for use by 44(mainly filesystem) devices and device I/O. This 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. The code implementing the 48VM Buffer abstraction is mostly concentrated in 49.Pa /usr/src/sys/kern/vfs_bio.c . 50.Pp 51One of the most important things to remember when dealing with buffer pointers 52(struct buf) is that the underlying pages are mapped directly from the buffer 53cache. No data copying occurs in the scheme proper, though some filesystems 54such as UFS do have to copy a little when dealing with file fragments. The 55second most important thing to remember is that due to the underlying page 56mapping, the b_data base pointer in a buf is always *page* aligned, not 57*block* aligned. When you have a VM buffer representing some b_offset and 58b_size, the actual start of the buffer is (b_data + (b_offset & PAGE_MASK)) 59and not just b_data. Finally, the VM system's core buffer cache supports 60valid and dirty bits (m->valid, m->dirty) for pages in DEV_BSIZE chunks. Thus 61a platform with a hardware page size of 4096 bytes has 8 valid and 8 dirty 62bits. These bits are generally set and cleared in groups based on the device 63block size of the device backing the page. Complete page's worth are often 64referred to using the VM_PAGE_BITS_ALL bitmask (i.e. 0xFF if the hardware page 65size is 4096). 66.Pp 67VM buffers also keep track of a byte-granular dirty range and valid range. 68This feature is normally only used by the NFS subsystem. I'm not sure why it 69is used at all, actually, since we have DEV_BSIZE valid/dirty granularity 70within the VM buffer. If a buffer dirty operation creates a 'hole', 71the dirty range will extend to cover the hole. If a buffer validation 72operation creates a 'hole' the byte-granular valid range is left alone and 73will not take into account the new extension. Thus the whole byte-granular 74abstraction is considered a bad hack and it would be nice if we could get rid 75of it completely. 76.Pp 77A VM buffer is capable of mapping the underlying VM cache pages into KVM in 78order to allow the kernel to directly manipulate the data associated with 79the (vnode,b_offset,b_size). The kernel typically unmaps VM buffers the moment 80they are no longer needed but often keeps the 'struct buf' structure 81instantiated and even bp->b_pages array instantiated despite having unmapped 82them from KVM. If a page making up a VM buffer is about to undergo I/O, the 83system typically unmaps it from KVM and replaces the page in the b_pages[] 84array with a placemarker called bogus_page. The placemarker forces any kernel 85subsystems referencing the associated struct buf to re-lookup the associated 86page. I believe the placemarker hack is used to allow sophisticated devices 87such as filesystem devices to remap underlying pages in order to deal with, 88for example, remapping a file fragment into a file block. 89.Pp 90VM buffers are used to track I/O operations within the kernel. Unfortunately, 91the I/O implementation is also somewhat of a hack because the kernel wants 92to clear the dirty bit on the underlying pages the moment it queues the I/O 93to the VFS device, not when the physical I/O is actually initiated. This 94can create confusion within filesystem devices that use delayed-writes because 95you wind up with pages marked clean that are actually still dirty. If not 96treated carefully, these pages could be thrown away! Indeed, a number of 97serious bugs related to this hack were not fixed until the 2.2.8/3.0 release. 98The kernel uses an instantiated VM buffer (i.e. struct buf) to placemark pages 99in this special state. The buffer is typically flagged B_DELWRI. When a 100device no longer needs a buffer it typically flags it as B_RELBUF. Due to 101the underlying pages being marked clean, the B_DELWRI|B_RELBUF combination must 102be interpreted to mean that the buffer is still actually dirty and must be 103written to its backing store before it can actually be released. In the case 104where B_DELWRI is not set, the underlying dirty pages are still properly 105marked as dirty and the buffer can be completely freed without losing that 106clean/dirty state information. ( XXX do we have to check other flags in 107regards to this situation ??? ). 108.Pp 109The kernel reserves a portion of its KVM space to hold VM Buffer's data 110maps. Even though this is virtual space (since the buffers are mapped 111from the buffer cache), we cannot make it arbitrarily large because 112instantiated VM Buffers (struct buf's) prevent their underlying pages in the 113buffer cache from being freed. This can complicate the life of the paging 114system. 115.Pp 116.\" .Sh SEE ALSO 117.\" .Xr <fillmein> 9 118.Sh HISTORY 119The 120.Nm 121manual page was originally written by 122.An Matthew Dillon 123and first appeared in 124.Fx 3.1 , 125December 1998. 126