xref: /netbsd-src/share/man/man9/mbuf.9 (revision bdc22b2e01993381dcefeff2bc9b56ca75a4235c)
1.\"	$NetBSD: mbuf.9,v 1.64 2018/05/07 10:53:45 maxv Exp $
2.\"
3.\" Copyright (c) 1997 The NetBSD Foundation, Inc.
4.\" All rights reserved.
5.\"
6.\" This documentation is derived from text contributed to The NetBSD Foundation
7.\" by S.P.Zeidler (aka stargazer).
8.\"
9.\" Redistribution and use in source and binary forms, with or without
10.\" modification, are permitted provided that the following conditions
11.\" are met:
12.\" 1. Redistributions of source code must retain the above copyright
13.\"    notice, this list of conditions and the following disclaimer.
14.\" 2. Redistributions in binary form must reproduce the above copyright
15.\"    notice, this list of conditions and the following disclaimer in the
16.\"    documentation and/or other materials provided with the distribution.
17.\"
18.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
19.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
20.\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
21.\" PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
22.\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
23.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
24.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
25.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
26.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
27.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
28.\" POSSIBILITY OF SUCH DAMAGE.
29.\"
30.Dd May 7, 2018
31.Dt MBUF 9
32.Os
33.Sh NAME
34.Nm mbuf ,
35.Nm m_get ,
36.Nm m_gethdr ,
37.Nm m_devget ,
38.Nm m_copym ,
39.Nm m_copypacket ,
40.Nm m_copydata ,
41.Nm m_copyback ,
42.Nm m_copyback_cow ,
43.Nm m_cat ,
44.Nm m_dup ,
45.Nm m_makewritable ,
46.Nm m_pulldown ,
47.Nm m_pullup ,
48.Nm m_copyup ,
49.Nm m_split ,
50.Nm m_adj ,
51.Nm m_apply ,
52.Nm m_free ,
53.Nm m_freem ,
54.Nm mtod ,
55.Nm MGET ,
56.Nm MGETHDR ,
57.Nm MEXTMALLOC ,
58.Nm MEXTADD ,
59.Nm MCLGET ,
60.Nm M_COPY_PKTHDR ,
61.Nm M_MOVE_PKTHDR ,
62.Nm m_remove_pkthdr ,
63.Nm M_ALIGN ,
64.Nm MH_ALIGN ,
65.Nm M_LEADINGSPACE ,
66.Nm M_TRAILINGSPACE ,
67.Nm M_PREPEND ,
68.Nm MCHTYPE
69.Nd "functions and macros for managing memory used by networking code"
70.Sh SYNOPSIS
71.In sys/mbuf.h
72.Ft struct mbuf *
73.Fn m_get "int how" "int type"
74.Ft struct mbuf *
75.Fn m_gethdr "int how" "int type"
76.Ft struct mbuf *
77.Fn m_devget "char *buf" "int totlen" "int off0" "struct ifnet *ifp" "void (*copy)(const void *, void *, size_t)"
78.Ft struct mbuf *
79.Fn m_copym "struct mbuf *m" "int off" "int len" "int wait"
80.Ft struct mbuf *
81.Fn m_copypacket "struct mbuf *m" "int how"
82.Ft void
83.Fn m_copydata "struct mbuf *m" "int off" "int len" "void *cp"
84.Ft void
85.Fn m_copyback "struct mbuf *m0" "int off" "int len" "void *cp"
86.Ft struct mbuf *
87.Fn m_copyback_cow "struct mbuf *m0" "int off" "int len" "void *cp" "int how"
88.Ft int
89.Fn m_makewritable "struct mbuf **mp" "int off" "int len" "int how"
90.Ft void
91.Fn m_cat "struct mbuf *m" "struct mbuf *n"
92.Ft struct mbuf *
93.Fn m_dup "struct mbuf *m" "int off" "int len" "int wait"
94.Ft struct mbuf *
95.Fn m_pulldown "struct mbuf *m" "int off" "int len" "int *offp"
96.Ft struct mbuf *
97.Fn m_pullup "struct mbuf *n" "int len"
98.Ft struct mbuf *
99.Fn m_copyup "struct mbuf *m" "int len" "int dstoff"
100.Ft struct mbuf *
101.Fn m_split "struct mbuf *m0" "int len" "int wait"
102.Ft void
103.Fn m_adj "struct mbuf *mp" "int req_len"
104.Ft int
105.Fn m_apply "struct mbuf *m" "int off" "int len" "int *f(void *, void *, unsigned int)" "void *arg"
106.Ft struct mbuf *
107.Fn m_free "struct mbuf *m"
108.Ft void
109.Fn m_freem "struct mbuf *m"
110.Ft datatype
111.Fn mtod "struct mbuf *m" "datatype"
112.Ft void
113.Fn MGET "struct mbuf *m" "int how" "int type"
114.Ft void
115.Fn MGETHDR "struct mbuf *m" "int how" "int type"
116.Ft void
117.Fn MEXTMALLOC "struct mbuf *m" "int len" "int how"
118.Ft void
119.Fn MEXTADD "struct mbuf *m" "void *buf" "int size" "int type" "void (*free)(struct mbuf *, void *, size_t, void *)" "void *arg"
120.Ft void
121.Fn MCLGET "struct mbuf *m" "int how"
122.Ft void
123.Fn M_COPY_PKTHDR "struct mbuf *to" "struct mbuf *from"
124.Ft void
125.Fn M_MOVE_PKTHDR "struct mbuf *to" "struct mbuf *from"
126.Ft void
127.Fn m_remove_pkthdr "struct mbuf *m"
128.Ft void
129.Fn M_ALIGN "struct mbuf *m" "int len"
130.Ft void
131.Fn MH_ALIGN "struct mbuf *m" "int len"
132.Ft int
133.Fn M_LEADINGSPACE "struct mbuf *m"
134.Ft int
135.Fn M_TRAILINGSPACE "struct mbuf *m"
136.Ft void
137.Fn M_PREPEND "struct mbuf *m" "int plen" "int how"
138.Ft void
139.Fn MCHTYPE "struct mbuf *m" "int type"
140.Sh DESCRIPTION
141The
142.Nm
143functions and macros provide an easy and consistent way to handle
144a networking stack's memory management needs.
145An
146.Nm
147consists of a header and a data area.
148It is of a fixed size,
149.Dv MSIZE
150.Pq defined in Aq Pa machine/param.h ,
151which includes the size of the header.
152The header contains a pointer to the next
153.Nm
154in the
155.Sq "mbuf chain" ,
156a pointer to the next
157.Sq "mbuf chain" ,
158a pointer to the data area, the amount of data in this mbuf, its type
159and a
160.Dv flags
161field.
162.Pp
163The
164.Dv type
165variable can signify:
166.Bl -tag -compact -offset indent -width "XXXXXXXXXXX"
167.It Dv MT_FREE
168the mbuf should be on the ``free'' list
169.It Dv MT_DATA
170data was dynamically allocated
171.It Dv MT_HEADER
172data is a packet header
173.It Dv MT_SONAME
174data is a socket name
175.It Dv MT_SOOPTS
176data is socket options
177.It Dv MT_FTABLE
178data is the fragment reassembly header
179.It Dv MT_CONTROL
180mbuf contains ancillary \&(protocol control\&) data
181.It Dv MT_OOBDATA
182mbuf contains out-of-band data.
183.El
184.Pp
185The
186.Dv flags
187variable contains information describing the
188.Nm ,
189notably:
190.Bl -tag -compact -offset indent -width "XXXXXXXXXXX"
191.It Dv M_EXT
192has external storage
193.It Dv M_PKTHDR
194is start of record
195.It Dv M_EOR
196is end of record
197.El
198.Pp
199If an
200.Nm
201designates the start of a record
202.Pq Dv M_PKTHDR ,
203its
204.Dv flags
205field may contain additional information describing the content of
206the record:
207.Bl -tag -compact -offset indent -width "XXXXXXXXXXX"
208.It Dv M_BCAST
209sent/received as link-level broadcast
210.It Dv M_MCAST
211sent/received as link-level multicast
212.El
213.Pp
214An
215.Nm
216may add a single
217.Sq "mbuf cluster"
218of
219.Dv MCLBYTES
220bytes
221.Pq defined in Aq Pa machine/param.h ,
222which has no additional overhead
223and is used instead of the internal data area; this is done when at least
224.Dv MINCLSIZE
225bytes of data must be stored.
226.Pp
227When the
228.Dv M_EXT
229flag is set on an mbuf,
230the external storage area could be shared among multiple mbufs.
231Therefore, care must be taken when overwriting the data content of an
232mbuf, because its external storage could be considered as read-only.
233.Bl -tag -width compact
234.It Fn m_get "int how" "int type"
235Allocates an mbuf and initializes it to contain internal data.
236The
237.Fa how
238parameter is a choice of
239.Dv M_WAIT / M_DONTWAIT
240from caller.
241.Dv M_WAIT
242means the call cannot fail, but may take forever.
243The
244.Fa type
245parameter is an mbuf type.
246.It Fn m_gethdr "int how" "int type"
247Allocates an mbuf and initializes it to contain a packet header and internal
248data.
249The
250.Fa how
251parameter is a choice of
252.Dv M_WAIT / M_DONTWAIT
253from caller.
254The
255.Fa type
256parameter is an mbuf type.
257.It Fn m_devget "char *buf" "int totlen" "int off0" "struct ifnet *ifp" "void (*copy)(const void *, void *, size_t)"
258Copies
259.Fa len
260bytes from device local memory into mbufs using copy routine
261.Fa copy .
262If parameter
263.Fa off
264is non-zero, the packet is supposed to be trailer-encapsulated and
265.Fa off
266bytes plus the type and length fields will be skipped before copying.
267Returns the top of the mbuf chain it created.
268.It Fn m_copym "struct mbuf *m" "int off" "int len" "int wait"
269Creates a copy of an mbuf chain starting
270.Fa off
271bytes from the beginning, continuing for
272.Fa len
273bytes.
274If the
275.Fa len
276requested is
277.Dv M_COPYALL ,
278the complete mbuf chain will be copied.
279The
280.Fa wait
281parameter is a choice of
282.Dv M_WAIT / M_DONTWAIT
283from caller.
284.It Fn m_copypacket "struct mbuf *m" "int how"
285Copies an entire packet, including header.
286This function is an optimization of the common case
287.Li m_copym ( m , 0 , Dv M_COPYALL , Fa how ) .
288However, contrary to
289.Fn m_copym ,
290a header must be present.
291It is incorrect to use
292.Fn m_copypacket
293with an mbuf that does not have a header.
294.It Fn m_copydata "struct mbuf *m" "int off" "int len" "void *cp"
295Copies
296.Fa len
297bytes data from mbuf chain
298.Fa m
299into the buffer
300.Fa cp ,
301starting
302.Fa off
303bytes from the beginning.
304.It Fn m_copyback "struct mbuf *m0" "int off" "int len" "void *cp"
305Copies
306.Fa len
307bytes data from buffer
308.Fa cp
309back into the mbuf chain
310.Fa m0 ,
311starting
312.Fa off
313bytes from the beginning of the chain, extending the mbuf chain if necessary.
314.Fn m_copyback
315can only fail when extending the chain.
316The caller should check for this kind of failure
317by checking the resulting length of the chain in that case.
318It is an error to use
319.Fn m_copyback
320on read-only mbufs.
321.It Fn m_copyback_cow "struct mbuf *m0" "int off" "int len" "void *cp" \
322"int how"
323Copies
324.Fa len
325bytes data from buffer
326.Fa cp
327back into the mbuf chain
328.Fa m0
329as
330.Fn m_copyback
331does.
332Unlike
333.Fn m_copyback ,
334it is safe to use
335.Fn m_copyback_cow
336on read-only mbufs.
337If needed,
338.Fn m_copyback_cow
339automatically allocates new mbufs and adjusts the chain.
340On success, it returns a pointer to the resulting mbuf chain,
341and frees the original mbuf
342.Fa m0 .
343Otherwise, it returns
344.Dv NULL .
345The
346.Fa how
347parameter is a choice of
348.Dv M_WAIT / M_DONTWAIT
349from the caller.
350Unlike
351.Fn m_copyback ,
352extending the mbuf chain isn't supported.
353It is an error to attempt to extend the mbuf chain using
354.Fn m_copyback_cow .
355.It Fn m_makewritable "struct mbuf **mp" "int off" "int len" "int how"
356Rearranges an mbuf chain so that
357.Fa len
358bytes from offset
359.Fa off
360are writable.
361When it meets read-only mbufs, it allocates new mbufs, adjusts the chain as
362.Fn m_copyback_cow
363does, and copies the original content into them.
364.Fn m_makewritable
365does
366.Em not
367guarantee that all
368.Fa len
369bytes at
370.Fa off
371are consecutive.
372The
373.Fa how
374parameter is a choice of
375.Dv M_WAIT / M_DONTWAIT
376from the caller.
377.Fn m_makewritable
378preserves the contents of the mbuf chain even in the case of failure.
379It updates a pointer to the mbuf chain pointed to by
380.Fa mp .
381It returns 0 on success.
382Otherwise, it returns an error code, typically
383.Er ENOBUFS .
384.It Fn m_cat "struct mbuf *m" "struct mbuf *n"
385Concatenates mbuf chain
386.Fa n
387to
388.Fa m .
389Both chains must be of the same type; packet headers will
390.Em not
391be updated if present.
392.It Fn m_dup "struct mbuf *m" "int off" "int len" "int wait"
393Similarly to
394.Fn m_copym ,
395the function creates a copy of an mbuf chain starting
396.Fa off
397bytes from the beginning, continuing for
398.Fa len
399bytes.
400While
401.Fn m_copym
402tries to share external storage for mbufs with
403.Dv M_EXT
404flag,
405.Fn m_dup
406will deep-copy the whole data content into new mbuf chain
407and avoids shared external storage.
408.It Fn m_pulldown "struct mbuf *m" "int off" "int len" "int *offp"
409Rearranges an mbuf chain so that
410.Fa len
411bytes from offset
412.Fa off
413are contiguous and in the data area of an mbuf.
414The return value points to an mbuf in the middle of the mbuf chain
415.Fa m .
416If we call the return value
417.Fa n ,
418the contiguous data region is available at
419.Li "mtod(n, void *) + *offp" ,
420or
421.Li "mtod(n, void *)"
422if
423.Fa offp
424is
425.Dv NULL .
426The top of the mbuf chain
427.Fa m ,
428and mbufs up to
429.Fa off ,
430will not be modified.
431On successful return, it is guaranteed that the mbuf pointed to by
432.Fa n
433does not have a shared external storage,
434therefore it is safe to update the contiguous region.
435Returns
436.Dv NULL
437and frees the mbuf chain on failure.
438.Fa len
439must be smaller than or equal to
440.Dv MCLBYTES .
441.It Fn m_pullup "struct mbuf *m" "int len"
442Rearranges an mbuf chain so that
443.Fa len
444bytes are contiguous
445and in the data area of an mbuf (so that
446.Fn mtod
447will work for a structure of size
448.Fa len ) .
449Returns the resulting
450mbuf chain on success, frees it and returns
451.Dv NULL
452on failure.
453If there is room, it will add up to
454.Dv max_protohdr
455-
456.Fa len
457extra bytes to the
458contiguous region to possibly avoid being called again.
459.Fa len
460must be smaller or equal than
461.Dv MHLEN .
462.It Fn m_copyup "struct mbuf *m" "int len" "int dstoff"
463Similar to
464.Fn m_pullup
465but copies
466.Fa len
467bytes of data into a new mbuf at
468.Fa dstoff
469bytes into the mbuf.
470The
471.Fa dstoff
472argument aligns the data and leaves room for a link layer header.
473Returns the new
474mbuf chain on success, and frees the mbuf chain and returns
475.Dv NULL
476on failure.
477Note that
478the function does not allocate mbuf clusters, so
479.Fa len + dstoff
480must be less than
481.Dv MHLEN .
482.It Fn m_split "struct mbuf *m0" "int len" "int wait"
483Partitions an mbuf chain in two pieces, returning the tail,
484which is all but the first
485.Fa len
486bytes.
487In case of failure, it returns
488.Dv NULL
489and restores the chain to its original state.
490.It Fn m_adj "struct mbuf *mp" "int req_len"
491Shaves off
492.Fa req_len
493bytes from head or tail of the (valid) data area.
494If
495.Fa req_len
496is greater than zero, front bytes are being shaved off, if it's smaller,
497from the back (and if it is zero, the mbuf will stay bearded).
498This function does not move data in any way, but is used to manipulate the
499data area pointer and data length variable of the mbuf in a non-clobbering
500way.
501.It Fn m_apply "struct mbuf *m" "int off" "int len" "int (*f)(void *, void *, unsigned int)" "void *arg"
502Apply function
503.Fa f
504to the data in an mbuf chain starting
505.Fa off
506bytes from the beginning, continuing for
507.Fa len
508bytes.
509Neither
510.Fa off
511nor
512.Fa len
513may be negative.
514.Fa arg
515will be supplied as first argument for
516.Fa f ,
517the second argument will be the pointer to the data buffer of a
518packet (starting after
519.Fa off
520bytes in the stream), and the third argument is the amount
521of data in bytes in this call.
522If
523.Fa f
524returns something not equal to zero
525.Fn m_apply
526will bail out, returning the return code of
527.Fa f .
528Upon successful completion it will return zero.
529.It Fn m_free "struct mbuf *m"
530Frees mbuf
531.Fa m .
532.It Fn m_freem "struct mbuf *m"
533Frees the mbuf chain beginning with
534.Fa m .
535This function contains the elementary sanity check for a
536.Dv NULL
537pointer.
538.It Fn mtod "struct mbuf *m" "datatype"
539Returns a pointer to the data contained in the specified mbuf
540.Fa m ,
541type-casted to the specified data type
542.Fa datatype .
543.It Fn MGET "struct mbuf *m" "int how" "int type"
544Allocates mbuf
545.Fa m
546and initializes it to contain internal data.
547See
548.Fn m_get .
549.It Fn MGETHDR "struct mbuf *m" "int how" "int type"
550Allocates mbuf
551.Fa m
552and initializes it to contain a packet header.
553See
554.Fn m_gethdr .
555.It Fn MEXTMALLOC "struct mbuf *m" "int len" "int how"
556Allocates external storage of size
557.Fa len
558for mbuf
559.Fa m .
560The
561.Fa how
562parameter is a choice of
563.Dv M_WAIT / M_DONTWAIT
564from caller.
565The flag
566.Dv M_EXT
567is set upon success.
568.It Fn MEXTADD "struct mbuf *m" "void *buf" "int size" "int type" "void (*free)(struct mbuf *, void *, size_t, void *)" "void *arg"
569Adds pre-allocated external storage
570.Fa buf
571to a normal mbuf
572.Fa m ;
573the parameters
574.Fa size ,
575.Fa type ,
576.Fa free
577and
578.Fa arg
579describe the external storage.
580.Fa size
581is the size of the storage,
582.Fa type
583describes its
584.Xr malloc 9
585type,
586.Fa free
587is a free routine (if not the usual one), and
588.Fa arg
589is a possible argument to the free routine.
590The flag
591.Dv M_EXT
592is set upon success.
593If a free routine is specified, it will be called when the mbuf is freed.
594In the case of former, the first argument for a free routine is the mbuf
595.Fa m
596and the routine is expected to free it in addition to the external storage
597pointed by second argument.
598In the case of latter, the first argument for the routine is NULL.
599.It Fn MCLGET "struct mbuf *m" "int how"
600Allocates and adds an mbuf cluster to a normal mbuf
601.Fa m .
602The
603.Fa how
604parameter is a choice of
605.Dv M_WAIT / M_DONTWAIT
606from caller.
607The flag
608.Dv M_EXT
609is set upon success.
610.It Fn M_COPY_PKTHDR "struct mbuf *to" "struct mbuf *from"
611Copies the mbuf pkthdr from mbuf
612.Fa from
613to mbuf
614.Fa to .
615.Fa from
616must have the type flag
617.Dv M_PKTHDR
618set, and
619.Fa to
620must be empty.
621.It Fn M_MOVE_PKTHDR "struct mbuf *to" "struct mbuf *from"
622Moves the mbuf pkthdr from mbuf
623.Fa from
624to mbuf
625.Fa to .
626.Fa from
627must have the type flag
628.Dv M_PKTHDR
629set, and
630.Fa to
631must be empty.
632The flag
633.Dv M_PKTHDR
634in mbuf
635.Fa from
636will be cleared.
637.It Fn m_remove_pkthdr "struct mbuf *m"
638Removes the mbuf pkthdr from mbuf
639.Fa m .
640.Fa m
641must have the flag
642.Dv M_PKTHDR
643set.
644This flag will be cleared.
645.It Fn M_ALIGN "struct mbuf *m" "int len"
646Sets the data pointer of a newly allocated mbuf
647.Fa m
648to
649.Fa len
650bytes from the end of the mbuf data area, so that
651.Fa len
652bytes of data written to the mbuf
653.Fa m ,
654starting at the data pointer, will be aligned to the end of the data area.
655.It Fn MH_ALIGN "struct mbuf *m" "int len"
656Sets the data pointer of a newly allocated packetheader mbuf
657.Fa m
658to
659.Fa len
660bytes from the end of the mbuf data area, so that
661.Fa len
662bytes of data written to the mbuf
663.Fa m ,
664starting at the data pointer, will be aligned to the end of the data area.
665.It Fn M_LEADINGSPACE "struct mbuf *m"
666Returns the amount of space available before the current start of valid
667data in mbuf
668.Fa m .
669Returns 0 if the mbuf data part is shared across multiple mbufs
670.Pq i.e. not writable .
671.It Fn M_TRAILINGSPACE "struct mbuf *m"
672Returns the amount of space available after the current end of valid
673data in mbuf
674.Fa m .
675Returns 0 if the mbuf data part is shared across multiple mbufs
676.Pq i.e. not writable .
677.It Fn M_PREPEND "struct mbuf *m" "int plen" "int how"
678Prepends space of size
679.Fa plen
680to mbuf
681.Fa m .
682If a new mbuf must be allocated,
683.Fa how
684specifies whether to wait.
685If
686.Fa how
687is
688.Dv M_DONTWAIT
689and allocation fails, the original mbuf chain is freed and
690.Fa m
691is set to
692.Dv NULL .
693It is illegal for the
694.Fa plen
695parameter to be greater than
696.Dv MHLEN .
697.It Fn MCHTYPE "struct mbuf *m" "int type"
698Change mbuf
699.Fa m
700to new type
701.Fa type .
702.El
703.Sh CODE REFERENCES
704The
705.Nm
706management functions are implemented within the file
707.Pa sys/kern/uipc_mbuf.c .
708Function prototypes, and the functions implemented as macros
709are located in
710.Pa sys/sys/mbuf.h .
711.Sh SEE ALSO
712.Pa /usr/share/doc/smm/18.net ,
713.Xr netstat 1 ,
714.Xr m_tag 9 ,
715.Xr malloc 9
716.Rs
717.%A Jun-ichiro Hagino
718.%T "Mbuf issues in 4.4BSD IPv6/IPsec support (experiences from KAME IPv6/IPsec implementation)"
719.%B "Proceedings of the freenix track: 2000 USENIX annual technical conference"
720.%D June 2000
721.Re
722.Sh AUTHORS
723.An -nosplit
724The original mbuf data structures were designed by Rob Gurwitz
725when he did the initial TCP/IP implementation at BBN.
726.Pp
727Further extensions and enhancements were made by Bill Joy, Sam Leffler,
728and Mike Karels at CSRG.
729.Pp
730Current implementation of external storage by
731.An Matt Thomas
732.Aq matt@3am-software.com
733and
734.An Jason R. Thorpe
735.Aq thorpej@NetBSD.org .
736