--- /dev/null
+.\" @(#)nhfsstone.man 1.13 89/10/05 Copyright (c) 1989, Legato Systems Inc
+.\" See DISCLAIMER file for restrictions
+.\"
+.\" Japanese Version Copyright (c) 2006 Yuichi SATO
+.\" all rights reserved.
+.\" Translated Wed Jun 14 06:50:33 JST 2006
+.\" by Yuichi SATO <ysato444@yahoo.co.jp> catch up to v1.0.7
+.\"
+.TH NHFSSTONE 8 "4 October 1989"
+.\"O .SH NAME
+.SH ̾Á°
+.\"O nhfsstone \- Network File System benchmark program
+nhfsstone \- ¥Í¥Ã¥È¥ï¡¼¥¯¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¤Î¥Ù¥ó¥Á¥Þ¡¼¥¯¥×¥í¥°¥é¥à
+.\"O .SH SYNOPSIS
+.SH ½ñ¼°
+.B nhfsstone
+[
+.B \-v
+] [[
+.B \-t secs
+] | [
+.B -c calls
+]] [
+.B \-l load
+] [
+.B \-p nprocs
+] [
+.B \-m mixfile
+] [
+.B dir
+]...
+.\"O .SH DESCRIPTION
+.SH ÀâÌÀ
+.\"O .B nhfsstone
+.\"O (pronounced n\-f\-s\-stone, the "h" is silent)
+.\"O is used on a
+.\"O .SM NFS
+.\"O client to generate an artificial load with a particular mix of
+.\"O .SM NFS
+.\"O operations. It reports the average response time of the server in
+.\"O milliseconds per call and the load in calls per second.
+.B nhfsstone
+(n\-f\-s\-stone ¤Èȯ²»¤·¡¢"h" ¤ÏÆɤޤʤ¤) ¤Ï
+.SM NFS
+¥¯¥é¥¤¥¢¥ó¥È¾å¤Ç¿Í°ÙŪ¤ÊÉé²Ù¤òÀ¸À®¤¹¤ë¤¿¤á¤Ë»ÈÍѤ¹¤ë¡£
+¤³¤ÎÉé²Ù¤Ë¤Ï¡¢ÆÃÄê¤ÎÁȤ߹ç¤ï¤»¤Î
+.SM NFS
+Áàºî¤¬´Þ¤Þ¤ì¤ë¡£
+¤³¤Î¥×¥í¥°¥é¥à¤Ï¥µ¡¼¥Ð¡¼¤ÎÊ¿¶Ñ±þÅú»þ´Ö (1 ²ó¤Î¸Æ¤Ó½Ð¤·Åö¤ê¤Î¥ß¥êÉÿô) ¤È
+Éé²Ù (1 ÉÃÅö¤ê¤Î¸Æ¤Ó½Ð¤·²ó¿ô) ¤òɽ¼¨¤¹¤ë¡£
+.\"O The program adjusts its calling patterns based on the client's kernel
+.\"O .SM NFS
+.\"O statistics and the elapsed time.
+¤³¤Î¥×¥í¥°¥é¥à¤Ï¡¢¥¯¥é¥¤¥¢¥ó¥È¤Î¥«¡¼¥Í¥ë¤Î
+.SM NFS
+Åý·×¤È½êÍ×»þ´Ö¤Ë´ð¤Å¤¤¤Æ¡¢¸Æ¤Ó½Ð¤·¥Ñ¥¿¡¼¥ó¤òÄ´À°¤¹¤ë¡£
+.\"O Load can be generated over a given time or number of
+.\"O .SM NFS
+.\"O calls.
+Éé²Ù¤Ï»þ´Ö¤Þ¤¿¤Ï
+.SM NFS
+¸Æ¤Ó½Ð¤·¤Î¿ô¤ò»ØÄꤹ¤ë¤³¤È¤ÇÀ¸À®¤Ç¤¤ë¡£
+.LP
+.\"O Because it uses the kernel
+.\"O .SM NFS
+.\"O statistics to monitor its progress,
+.\"O .B nhfsstone
+.\"O cannot be used to measure the performance of non\-NFS filesystems.
+.B nhfsstone
+¤Ï¿ÊĽ¾õ¶·¤Î´Æ»ë¤Ë¥«¡¼¥Í¥ë¤Î
+.SM NFS
+Åý·×¾ðÊó¤ò»ÈÍѤ¹¤ë¤Î¤Ç¡¢
+NFS ¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à°Ê³°¤ÎÀǽ¤ò·×¬¤¹¤ë¤Î¤Ë¤Ï»È¤¨¤Ê¤¤¡£
+.LP
+.\"O The
+.\"O .B nhfsstone
+.\"O program uses file and directory manipulation in an attempt to generate
+.\"O particular
+.\"O .SM NFS
+.\"O operations in response to particular system calls.
+.B nhfsstone
+¥×¥í¥°¥é¥à¤Ï¡¢¥Õ¥¡¥¤¥ë¤È¥Ç¥£¥ì¥¯¥È¥êÁàºî¤ò»È¤Ã¤Æ¡¢
+ÆÃÄê¤Î¥·¥¹¥Æ¥à¥³¡¼¥ë¤ËÂбþ¤¹¤ë
+.SM NFS
+Áàºî¤òÀ¸À®¤¹¤ë¡£
+.\"O To do this it uses several tricks
+.\"O that are based on a knowledge of the implementation of the
+.\"O .SM NFS
+.\"O client side reference port.
+¤³¤ì¤ò¹Ô¤¦¤¿¤á¡¢
+.SM NFS
+¥¯¥é¥¤¥¢¥ó¥È¦¤Î»²¾È¥Ý¡¼¥È¤Î¼ÂÁõ¤Ë¤Ä¤¤¤Æ¤ÎÃμ±¤Ë´ð¤Å¤¤¤¿¡¢
+¤¤¤¯¤Ä¤«¤Î¥È¥ê¥Ã¥¯¤ò»È¤Ã¤Æ¤¤¤ë¡£
+.\"O For example, it uses long file names to circumvent the kernel name lookup
+.\"O cache so that a
+.\"O .BR stat (2)
+.\"O system call generates an
+.\"O .SM NFS
+.\"O lookup operation.
+Î㤨¤Ð¡¢Ä¹¤¤¥Õ¥¡¥¤¥ë̾¤ò»È¤Ã¤Æ¥«¡¼¥Í¥ë¤Î̾Á°¸¡º÷¥¥ã¥Ã¥·¥å¤ò²óÈò¤µ¤»¤ë¤³¤È¤Ç¡¢
+.BR stat (2)
+¥·¥¹¥Æ¥à¥³¡¼¥ë¤Ç
+.SM NFS
+lookup Áàºî¤òÀ¸À®¤¹¤ë¡£
+.LP
+.\"O The mix of
+.\"O .SM NFS
+.\"O operations can be set with a mix file, which is the output of the
+.\"O .BR nfsstat (8C)
+.\"O command (see the "\-m" option below).
+.SM NFS
+Áàºî¤ÎÁȤ߹ç¤ï¤»¤Ï¡¢ÁȤ߹ç¤ï¤» (mix) ¥Õ¥¡¥¤¥ë¤ÇÀßÄê¤Ç¤¤ë¡£
+¤³¤ì¤Ï
+.BR nfsstat (8C)
+¥³¥Þ¥ó¥É (²¼µ¤Î "\-m" ¥ª¥×¥·¥ç¥ó¤ò»²¾È) ¤Î½ÐÎϤǤ¢¤ë¡£
+.\"O The percentages taken from
+.\"O the mix file are calculated based on the number of
+.\"O .SM NFS
+.\"O calls, not on the percentages printed by nfsstat. Operations with
+.\"O 0% in the mix will never get called by
+.\"O .BR nhfsstone .
+ÁȤ߹ç¤ï¤»¥Õ¥¡¥¤¥ë¤«¤é¼èÆÀ¤µ¤ì¤ë³ä¹ç¤Ï¡¢
+nfsstat
+¤Çɽ¼¨¤µ¤ì¤ë³ä¹ç¤Ç¤Ï¤Ê¤¯¡¢
+.SM NFS
+¸Æ¤Ó½Ð¤·¤Î¿ô¤Ë´ð¤Å¤¤¤Æ·×»»¤µ¤ì¤ë¡£
+ÁȤ߹ç¤ï¤»¥Õ¥¡¥¤¥ë¤Ç 0% ¤È¤Ê¤Ã¤Æ¤¤¤ëÁàºî¤Ï¡¢
+.B nhfsstone
+¤«¤é¸Æ¤Ó½Ð¤µ¤ì¤Ê¤¤¡£
+.\"O In a real server load mix, even though the percentage of call for
+.\"O a particular
+.\"O .SM NFS
+.\"O operation may be zero, the number of calls is often nonzero.
+¼ÂºÝ¤Î¥µ¡¼¥Ð¡¼¤Ë¤ª¤±¤ëÉé²Ù¤ÎÁȤ߹ç¤ï¤»¤Ç¤Ï¡¢ÆÃÄê¤Î
+.SM NFS
+Áàºî¤Î³ä¹ç¤¬ 0% ¤Ë¤Ê¤Ã¤Æ¤¤¤ë¤«¤â¤·¤ì¤Ê¤¤¤¬¡¢
+¸Æ¤Ó½Ð¤·¿ô¤Ï 0 ¤Ç¤Ê¤¤¤³¤È¤¬Â¿¤¤¡£
+.\"O .B Nhfsstone
+.\"O makes the assumption that the number of calls to these 0 percent
+.\"O operations will have an insignificant effect on server response.
+.B nhfsstone
+¤Ï¡Ö³ä¹ç¤¬ 0% ¤Ë¤Ê¤Ã¤Æ¤¤¤ëÁàºî¤Î¸Æ¤Ó½Ð¤·¿ô¤Ï¡¢
+¥µ¡¼¥Ð¡¼¤Î±þÅú¤Ëº³ºÙ¤Ê±Æ¶Á¤·¤«Í¿¤¨¤Ê¤¤¡×¤È²¾Äꤷ¤Æ¤¤¤ë¡£
+.LP
+.\"O Normally
+.\"O .B nhfsstone
+.\"O should be given a list of two or more test directories to use
+.\"O (default is to use the current directory).
+Ä̾ï¤Ï
+.B nhfsstone
+¤ËÂФ·¤Æ¡¢»ÈÍѤ¹¤ë 2 ¤Ä°Ê¾å¤Î¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤Î¥ê¥¹¥È¤ò»ØÄꤹ¤Ù¤¤Ç¤¢¤ë
+(¥Ç¥Õ¥©¥ë¥È¤Ç¤Ï¥«¥ì¥ó¥È¥Ç¥£¥ì¥¯¥È¥ê¤ò»ÈÍѤ¹¤ë)¡£
+.\"O The test directories used should be located on different disks and
+.\"O partitions on the server to realistically simulate typical server loads.
+»ÈÍѤ¹¤ë¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤Ï¡¢
+ŵ·¿Åª¤Ê¥µ¡¼¥Ð¡¼¤ÎÉé²Ù¤ò¸½¼ÂŪ¤Ë¥·¥ß¥å¥ì¡¼¥È¤¹¤ë¤¿¤á¤Ë¡¢
+¥µ¡¼¥Ð¡¼¾å¤ÎÊÌ¡¹¤Î¥Ç¥£¥¹¥¯¤È¥Ñ¡¼¥Æ¥£¥·¥ç¥ó¤ËÃÖ¤¯¤Ù¤¤Ç¤¢¤ë¡£
+.\"O Each
+.\"O .B nhfsstone
+.\"O process looks for a directory
+.\"O .B <dir>/testdir<n>
+.\"O (where <n> is a number from 0 to
+.\"O .B nprocs
+.\"O \- 1).
+³Æ
+.B nhfsstone
+¥×¥í¥»¥¹¤Ï¥Ç¥£¥ì¥¯¥È¥ê
+.B <dir>/testdir<n>
+¤ò°·¤¦
+(¤³¤³¤Ç <n> ¤Ï 0 ¤«¤é
+.B nprocs
+\- 1 ¤Þ¤Ç¤Î¿ôÃͤǤ¢¤ë)¡£
+.\"O If a process directory name already exists,
+.\"O it is checked for the correct set of test files.
+.\"O Otherwise the directory is created and populated.
+½èÍý¤¹¤ë¥Ç¥£¥ì¥¯¥È¥ê̾¤¬´û¤Ë¸ºß¤¹¤ë¾ì¹ç¡¢
+¥Æ¥¹¥È¥Õ¥¡¥¤¥ë¤Î¥»¥Ã¥È¤¬Àµ¤·¤¤¤«¤ò¥Á¥§¥Ã¥¯¤¹¤ë¡£
+¸ºß¤·¤Ê¤¤¾ì¹ç¤Ï¡¢¥Ç¥£¥ì¥¯¥È¥ê¤òºîÀ®¤·¤Æ (ÌõÃð: ¥Æ¥¹¥È¥Õ¥¡¥¤¥ë¤ò) ÇÛÃÖ¤¹¤ë¡£
+.\"O .SH OPTIONS
+.SH ¥ª¥×¥·¥ç¥ó
+.TP 12
+.B \-v
+.\"O Verbose output.
+¾ÜºÙ¤Ê½ÐÎÏ¡£
+.TP
+.B \-t secs
+.\"O Sets
+.\"O .B calls
+.\"O based on the given running time (in seconds) and the load.
+»ØÄꤷ¤¿¼Â¹Ô»þ´Ö (ñ°Ì: ÉÃ) ¤ÈÉé²Ù¤Ë´ð¤Å¤¤¤Æ¡¢
+.B calls
+¤òÀßÄꤹ¤ë¡£
+.TP
+.B \-c calls
+.\"O Total number of
+.\"O .SM NFS
+.\"O calls to generate (default is 5000).
+À¸À®¤¹¤ë
+.SM NFS
+¸Æ¤Ó½Ð¤·¤ÎÁí¿ô (¥Ç¥Õ¥©¥ë¥È¤Ï 5000)¡£
+.TP
+.B \-l load
+.\"O Load to generate in
+.\"O .SM NFS
+.\"O calls per second (default is 30).
+1 ÉÃÅö¤ê¤Î
+.SM NFS
+¸Æ¤Ó½Ð¤·¤òÀ¸À®¤¹¤ëÉé²Ù (¥Ç¥Õ¥©¥ë¥È¤Ï 30)¡£
+.TP
+.B \-p nprocs
+.\"O Number of load generating sub\-processes to fork (default is 7).
+.\"O This can be used to maximize the amount of load a single machine can generate.
+.\"O On a small client machine (slow CPU or small amount of memory)
+.\"O fewer processes might be used to avoid swapping.
+Éé²ÙÀ¸À®¤ò¹Ô¤¦¥µ¥Ö¥×¥í¥»¥¹¤ò fork ¤µ¤»¤ë¿ô (¥Ç¥Õ¥©¥ë¥È¤Ï 7)¡£
+¤³¤ì¤Ï 1 ¤Ä¤Î¥Þ¥·¥ó¤ÇÀ¸À®¤Ç¤¤ëÉé²Ù¤ÎÁíÎ̤òºÇÂç²½¤¹¤ë¤¿¤á¤Ë»ÈÍѤµ¤ì¤ë¡£
+(CPU ¤¬ÃÙ¤¤¡¢¤Þ¤¿¤Ï¥á¥â¥êÎ̤¬¾¯¤Ê¤¤) ¾®¤µ¤Ê¥¯¥é¥¤¥¢¥ó¥È¥Þ¥·¥ó¤Ç¤Ï¡¢
+¥¹¥ï¥Ã¥Ô¥ó¥°¤òÈò¤±¤ë¤¿¤á¤Ë¡¢¥×¥í¥»¥¹¿ô¤ò¾¯¤Ê¤¯¤·¤¿Êý¤¬Îɤ¤¤«¤â¤·¤ì¤Ê¤¤¡£
+.TP
+.B \-m mixfile
+.\"O Mix of
+.\"O .SM NFS
+.\"O operations to generate.
+À¸À®¤¹¤ë
+.SM NFS
+Áàºî¤ÎÁȤ߹ç¤ï¤»¡£
+.\"O The format of
+.\"O .B mixfile
+.\"O is the same as the output of the
+.\"O .BR nfsstat (8C)
+.\"O program.
+.B mixfile
+¤Î¥Õ¥©¡¼¥Þ¥Ã¥È¤Ï
+.BR nfsstat (8C)
+¥×¥í¥°¥é¥à¤Î½ÐÎϤÈƱ¤¸¤Ç¤¢¤ë¡£
+.\"O A mix file can be created on a server by typing "nfsstat \-s > mixfile".
+ÁȤ߹ç¤ï¤»¥Õ¥¡¥¤¥ë¤Ï¡¢¥µ¡¼¥Ð¡¼¾å¤Ç "nfsstat \-s > mixfile" ¤ò
+¼Â¹Ô¤¹¤ë¤³¤È¤ÇÀ¸À®¤Ç¤¤ë¡£
+.\"O The default mix of operations is: null 0%, getattr 13%, setattr 1%,
+.\"O root 0%, lookup 34%, readlink 8%, read 22%, wrcache 0%, write 15%, create 2%,
+.\"O remove 1%, rename 0%, link 0%, symlink 0%, mkdir 0%, rmdir 0%, readdir 3%,
+.\"O fsstat 1%.
+¥Ç¥Õ¥©¥ë¥È¤ÎÁàºî¤ÎÁȤ߹ç¤ï¤»¤Ï°Ê²¼¤ÎÄ̤ê:
+null 0%, getattr 13%, setattr 1%,
+root 0%, lookup 34%, readlink 8%, read 22%, wrcache 0%, write 15%, create 2%,
+remove 1%, rename 0%, link 0%, symlink 0%, mkdir 0%, rmdir 0%, readdir 3%,
+fsstat 1%.
+.\"O .SH USING NHFSSTONE
+.SH NHFSSTONE ¤Î»ÈÍÑË¡
+.\"O As with all benchmarks,
+.\"O .B nhfsstone
+.\"O can only provide numbers that are useful if experiments that use it are
+.\"O set up carefully.
+¾¤ÎÁ´¤Æ¤Î¥Ù¥ó¥Á¥Þ¡¼¥¯¤ÈƱÍͤˡ¢
+»ÈÍѤµ¤ì¤ë»î¸³ÆâÍƤòÃí°Õ¿¼¤¯ÀßÄꤷ¤¿¾ì¹ç¤Ë¤Î¤ß¡¢
+.B nhfsstone
+¤ÏÍÍѤʿôÃͤòÄ󶡤Ǥ¤ë¡£
+.\"O Since it is measuring servers, it should be run on a client
+.\"O that will not limit the generation of
+.\"O .SM NFS
+.\"O requests.
+¤³¤Î¥×¥í¥°¥é¥à¤Ï¥µ¡¼¥Ð¡¼¤ò¬Äꤹ¤ë¤Î¤Ç¡¢
+.SM NFS
+¥ê¥¯¥¨¥¹¥È¤ÎÀ¸À®¤ËÀ©¸Â¤¬¤Ê¤¤¥¯¥é¥¤¥¢¥ó¥È¾å¤Ç¼Â¹Ô¤¹¤Ù¤¤Ç¤¢¤ë¡£
+.\"O This means it should have a fast CPU,
+.\"O a good ethernet interface and the machine
+.\"O should not be used for anything else during testing.
+¥¯¥é¥¤¥¢¥ó¥È¥Þ¥·¥ó¤Ï¡¢Â®¤¤ CPU¤ÈÍ¥¤ì¤¿¥¤¡¼¥µ¡¼¥Í¥Ã¥È¥¤¥ó¥¿¥Õ¥§¡¼¥¹¤òÈ÷¤¨¡¢
+¥Æ¥¹¥È¤ÎºÇÃæ¤Ë¤½¤Î¾¤ÎÍÑÅӤǻÈÍѤ¹¤Ù¤¤Ç¤Ï¤Ê¤¤¡¢¤È¤¤¤¦°ÕÌ£¤Ç¤¢¤ë¡£
+.\"O A Sun\-3/50 can generate about 60
+.\"O .SM NFS
+.\"O calls per second before it runs out of CPU.
+Sun\-3/50 ¤Ç¤Ï¡¢CPU ¤ò»È¤¤¤¤ëÁ°¤Ë¡¢1 ÉÃÅö¤ê 60
+.SM NFS
+¸Æ¤Ó½Ð¤·¤¬²Äǽ¤Ç¤¢¤ë¡£
+.LP
+.\"O .B Nhfsstone
+.\"O assumes that all
+.\"O .SM NFS
+.\"O calls generated on the client are going to a single server, and that
+.\"O all of the
+.\"O .SM NFS
+.\"O load on that server is due to this client.
+.B nhfsstone
+¤Ï¡Ö¥¯¥é¥¤¥¢¥ó¥È¤ÇÀ¸À®¤µ¤ì¤¿Á´¤Æ¤Î
+.SM NFS
+¸Æ¤Ó½Ð¤·¤¬ 1 ¤Ä¤Î¥µ¡¼¥Ð¡¼¤ËÂФ·¤Æ¹Ô¤ï¤ì¡¢Á´¤Æ¤Î
+.SM NFS
+Éé²Ù¤Ï¤³¤Î¥¯¥é¥¤¥¢¥ó¥È¤Ë°Í¤ë¤â¤Î¤Ç¤¢¤ë¡×¤È²¾Äꤷ¤Æ¤¤¤ë¡£
+.\"O To make this assumption hold,
+.\"O both the client and server should be as quiescent as possible during tests.
+¤³¤Î²¾Äê¤ò°Ý»ý¤¹¤ë¤¿¤á¡¢¥Æ¥¹¥È¤ÎºÇÃæ¤Ï
+¥¯¥é¥¤¥¢¥ó¥È¤È¥µ¡¼¥Ð¡¼¤ÎξÊý¤ò½ÐÍè¤ë¸Â¤êÀŤ«¤Ë¤µ¤»¤Æ
+(¾¤Î»Å»ö¤ò¤µ¤»¤Ê¤¤¤Ç) ¤ª¤¯¤Ù¤¤Ç¤¢¤ë¡£
+.LP
+.\"O If the network is heavily utilized the delays due to collisions
+.\"O may hide any changes in server performance.
+.\"O High error rates on either the client or server can also
+.\"O cause delays due to retransmissions of lost or damaged packets.
+¥Í¥Ã¥È¥ï¡¼¥¯¤ò¹âÉé²Ù¤Ç»ÈÍѤ·¤Æ¤¤¤ë¤È¡¢
+¾×Æͤˤè¤ëÃ٤줬¥µ¡¼¥Ð¡¼Àǽ¤ÎÊѲ½¤ò±£¤·¤Æ¤·¤Þ¤¦¤«¤â¤·¤ì¤Ê¤¤¡£
+¥¯¥é¥¤¥¢¥ó¥È¤È¥µ¡¼¥Ð¡¼¤Î¤É¤Á¤é¤«¤Ç¥¨¥é¡¼Î¨¤¬¹â¤¤¤È¡¢
+¼º¤ï¤ì¤¿¥Ñ¥±¥Ã¥È¤ä²õ¤ì¤¿¥Ñ¥±¥Ã¥È¤òºÆÁ÷¿®¤¹¤ë¤¿¤á¤ËÃ٤줬À¸¤¸¤ë²ÄǽÀ¤¬¤¢¤ë¡£
+.\"O .BR netstat (8C)
+.\"O .B \-i
+.\"O can be used to measure the error and collision rates on the client and server.
+.BR netstat (8C)
+.B \-i
+¤ò»È¤Ã¤Æ¥¯¥é¥¤¥¢¥ó¥È¤È¥µ¡¼¥Ð¡¼¤Ë¤ª¤±¤ë¥¨¥é¡¼Î¨¤È¾×ÆÍΨ¤ò·×¬¤Ç¤¤ë¡£
+.LP
+.\"O To best simulate the effects of
+.\"O .SM NFS
+.\"O clients on the server, the test
+.\"O directories should be set up so that they are on at least two of the
+.\"O disk partitions that the server exports and the partitions should be
+.\"O as far apart as possible. The
+.\"O .BR dkinfo (8)
+.\"O command can be used to find the physical geometry of disk on BSD based systems.
+¥µ¡¼¥Ð¡¼¤ËÂФ¹¤ë
+.SM NFS
+¥¯¥é¥¤¥¢¥ó¥È¤Î±Æ¶Á¤òºÇ¤âÀµ¤·¤¯¥·¥ß¥å¥ì¡¼¥È¤¹¤ë¤Ë¤Ï¡¢
+¥µ¡¼¥Ð¡¼¤¬¥¨¥¯¥¹¥Ý¡¼¥È¤¹¤ë¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤ò
+¾¯¤Ê¤¯¤È¤â 2 ¤Ä¤Î¥Ç¥£¥¹¥¯¥Ñ¡¼¥Æ¥£¥·¥ç¥ó¤ËÃÖ¤¤¤Æ¡¢
+¤«¤Ä¤½¤ì¤é¤Î¥Ñ¡¼¥Æ¥£¥·¥ç¥ó¤ò½ÐÍè¤ë¸Â¤êÎ¥¤¹¤Ù¤¤Ç¤¢¤ë¡£
+.BR dkinfo (8)
+¥³¥Þ¥ó¥É¤ò»È¤¦¤È¡¢BSD ·Ï¥·¥¹¥Æ¥à¤Ë¤ª¤±¤ë
+¥Ç¥£¥¹¥¯¤ÎʪÍýŪ¥¸¥ª¥á¥È¥ê¤¬Ê¬¤«¤ë¡£
+.\"O .SM NFS
+.\"O operations tend to randomize
+.\"O access the whole disk so putting all of the
+.\"O .B nhfsstone
+.\"O test directories on a single partition or on
+.\"O two partitions that are close together will not show realistic results.
+.SM NFS
+Áàºî¤Ç¤Ï¥Ç¥£¥¹¥¯Á´ÂΤإé¥ó¥À¥à¤Ë¥¢¥¯¥»¥¹¤¹¤ë·¹¸þ¤¬¤¢¤ë¤Î¤Ç¡¢
+Á´¤Æ¤Î
+.B nhfsstone
+¤Î¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤ò 1 ¤Ä¤Î¥Ñ¡¼¥Æ¥£¥·¥ç¥ó¤ËÃÖ¤¤¤¿¤ê¡¢
+2 ¤Ä¤Î¥Ñ¡¼¥Æ¥£¥·¥ç¥ó¤¬¶á¤«¤Ã¤¿¤ê¤¹¤ë¤È¡¢¸½¼ÂŪ¤Ê·ë²Ì¤ò¼¨¤µ¤Ê¤¤¡£
+.LP
+.\"O On all tests it is a good idea to run the tests repeatedly and compare results.
+Á´¤Æ¤Î¥Æ¥¹¥È¤Ë¤ª¤¤¤Æ¡¢¥Æ¥¹¥È¤ò·«¤êÊÖ¤·¹Ô¤Ã¤Æ
+·ë²Ì¤òÈæ³Ó¤¹¤ë¤Î¤Ï¡¢Îɤ¤¹Í¤¨¤Ç¤¢¤ë¡£
+.\"O The number of calls can be increased
+.\"O (with the
+.\"O .B \-c
+.\"O option) until the variance in milliseconds per call is acceptably small.
+¸Æ¤Ó½Ð¤·¤Î¿ô¤Ï
+.RB ( \-c
+¥ª¥×¥·¥ç¥ó¤ò»È¤Ã¤Æ) ¸Æ¤Ó½Ð¤·°ì²óÅö¤ê¤Î»þ´Ö (¥ß¥êÉÃ) ¤ÎÊѲ½¤¬
+½¼Ê¬¾®¤µ¤¯¤Ê¤ë¤Þ¤ÇÁý¤ä¤¹¤³¤È¤¬¤Ç¤¤ë¡£
+.\"O If increasing the number of calls does not help there may be something
+.\"O wrong with the experimental setup.
+¸Æ¤Ó½Ð¤·²ó¿ô¤òÁý¤ä¤·¤Æ¤â½õ¤±¤Ë¤Ê¤é¤Ê¤¤ (ÌõÃð: ²¿¤ÎÊѲ½¤â¸«¤é¤ì¤Ê¤¤) ¾ì¹ç¡¢
+¼Â¸³¤ÎÀßÄê¤Ë²¿¤«ÌäÂ꤬¤¢¤ë¤«¤â¤·¤ì¤Ê¤¤¡£
+.\"O One common problem is too much memory on the client
+.\"O test machine. With too much memory,
+.\"O .B nhfsstone
+.\"O is not able to defeat the client caches and the
+.\"O .SM NFS
+.\"O operations do not end up going to the server at all. If you suspect that
+.\"O there is a caching problem you can use the
+.\"O .B -p
+.\"O option to increase the number of processes.
+1 ¤Ä¤Î°ìÈÌŪ¤ÊÌäÂê¤È¤·¤Æ¤Ï¡¢
+¥¯¥é¥¤¥¢¥ó¥È¤Î¥Æ¥¹¥È¥Þ¥·¥ó¤Ë¥á¥â¥ê¤¬Â¿¤¹¤®¤ë¾ì¹ç¤¬¤¢¤ë¡£
+¥á¥â¥ê¤¬Â¿¤¹¤®¤ë¤È¡¢
+.B nhfsstone
+¤¬¥¯¥é¥¤¥¢¥ó¥È¤Î¥¥ã¥Ã¥·¥å¤Ë¾¡¤Æ¤º¡¢
+.SM NFS
+Áàºî¤¬¥µ¡¼¥Ð¡¼¤ËÁ´¤¯¹Ô¤«¤Ê¤¯¤Ê¤ë¡£
+¥¥ã¥Ã¥·¥å¤ËÌäÂ꤬¤¢¤ë¤È´¶¤¸¤¿¾ì¹ç¤Ï¡¢
+.B -p
+¥ª¥×¥·¥ç¥ó¤ò»È¤Ã¤Æ¥×¥í¥»¥¹¿ô¤òÁý¤ä¤¹¤³¤È¤¬¤Ç¤¤ë¡£
+.LP
+.\"O The numbers generated by
+.\"O .B nhfsstone
+.\"O are most useful for comparison if the test setup on the client machine
+.\"O is the same between different server configurations.
+.B nhfsstone
+¤ÇÀ¸À®¤µ¤ì¤ë¿ôÃͤϡ¢¥¯¥é¥¤¥¢¥ó¥È¥Þ¥·¥ó¤Ç¤Î¥Æ¥¹¥ÈÀßÄê¤òÅù¤·¤¯¤·¤Æ¡¢
+°Û¤Ê¤ë¥µ¡¼¥Ð¡¼ÀßÄê¤òÈæ³Ó¤¹¤ë¾ì¹ç¤ËºÇ¤â͸ú¤Ç¤¢¤ë¡£
+.\"O Changing
+.\"O .B nhfsstone
+.\"O parameters between runs will produce numbers that can not be
+.\"O meaningfully compared.
+¼Â¹Ô¤¹¤ëËè¤Ë
+.B nhfsstone
+¤Î¥Ñ¥é¥á¡¼¥¿¤òÊѹ¹¤¹¤ë¤È¡¢°ÕÌ£¤Î¤¢¤ëÈæ³Ó¤¬½ÐÍè¤Ê¤¤¿ôÃͤ¬À¸À®¤µ¤ì¤ë¡£
+.\"O For example, changing the number of generator processes
+.\"O may affect the measured response
+.\"O time due to context switching or other delays on the client machine, while
+.\"O changing the mix of
+.\"O .SM NFS
+.\"O operations will change the whole nature of the experiment.
+¤¿¤È¤¨¤Ð¡¢Éé²Ù¤òÀ¸À®¤¹¤ë¥×¥í¥»¥¹¤Î¿ô¤Ï¡¢
+¥¯¥é¥¤¥¢¥ó¥È¥Þ¥·¥ó¤Ç¤Î¥³¥ó¥Æ¥¯¥¹¥ÈÀÚ¤êÂؤ¨¤ä¤½¤Î¾¤ÎÃÙ¤ì¤Ë¤è¤ê¡¢
+·×¬¤µ¤ì¤ë±þÅú»þ´Ö¤Ë±Æ¶Á¤òÍ¿¤¨¤ë¤«¤â¤·¤ì¤Ê¤¤¡£
+°ìÊý¡¢
+.SM NFS
+Áàºî¤ÎÁȤ߹ç¤ï¤»¤òÊѹ¹¤¹¤ë¤È¡¢¼Â¸³¤ÎÁ´ÂÎŪ¤ÊÆÃÀ¤òÊѤ¨¤ë¤³¤È¤Ë¤Ê¤ë¡£
+.\"O Other changes to the client configuration may also effect the comparability
+.\"O of results.
+¤½¤Î¾¤Î¥¯¥é¥¤¥¢¥ó¥ÈÀßÄê¤Ë¤Ä¤¤¤Æ¤ÎÊѹ¹¤Ï¡¢
+·ë²Ì¤¬Èæ³Ó¤Ç¤¤ë¤«¤Ë±Æ¶Á¤òÍ¿¤¨¤ë²ÄǽÀ¤¬¤¢¤ë¡£
+.\"O While
+.\"O .B nhfsstone
+.\"O tries to compensate for differences in client configurations
+.\"O by sampling the actual
+.\"O .SM NFS
+.\"O statistics and adjusting both the load and mix of operations, some changes
+.\"O are not reflected in either the load or the mix. For example, installing
+.\"O a faster CPU or mounting different
+.\"O .SM NFS
+.\"O filesystems may effect the response time without changing either the
+.\"O load or the mix.
+.B nhfsstone
+¤Ï¼ÂºÝ¤Î
+.SM NFS
+Åý·×¤ò¥µ¥ó¥×¥ê¥ó¥°¤·¤Æ¡¢Áàºî¤ÎÉé²Ù¤ÈÁȤ߹ç¤ï¤»¤òÄ´À°¤¹¤ë¤³¤È¤Ç¡¢
+¥¯¥é¥¤¥¢¥ó¥ÈÀßÄê¤Î°ã¤¤¤òÊäÀµ¤·¤è¤¦¤È¤¹¤ë¤¬¡¢
+¤¤¤¯¤Ä¤«¤ÎÊѹ¹¤ÏÉé²Ù¤Ë¤âÁȹç¤ï¤»¤Ë¤âÈ¿±Ç¤µ¤ì¤Ê¤¤¡£
+Î㤨¤Ð¡¢Éé²Ù¤âÁȤ߹ç¤ï¤»¤âÊѹ¹¤»¤º¡¢¤è¤ê®¤¤ CPU ¤òÁȤ߹þ¤ó¤À¤ê¡¢
+Ê̤Î
+.SM NFS
+¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¤ò¥Þ¥¦¥ó¥È¤¹¤ë¤È¡¢±þÅú»þ´Ö¤Ë±Æ¶Á¤¬½Ð¤ë¤À¤í¤¦¡£
+.LP
+.\"O To do a comparison of different server configurations, first set up the
+.\"O client test directories and do
+.\"O .B nhfsstone
+.\"O runs at different loads to be sure that the variability is
+.\"O reasonably low. Second, run
+.\"O .B nhfsstone
+.\"O at different loads of interest and
+.\"O save the results. Third, change the server configuration (for example,
+.\"O add more memory, replace a disk controller, etc.). Finally, run the same
+.\"O .B nhfsstone
+.\"O loads again and compare the results.
+°Û¤Ê¤ë¥µ¡¼¥Ð¡¼ÀßÄê¤òÈæ³Ó¤¹¤ë¾ì¹ç¡¢
+Âè 1 Ãʳ¬¤Ç¤Ï¡¢¥¯¥é¥¤¥¢¥ó¥È¤Î¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤òÀßÄꤷ¡¢
+.B nhfsstone
+¤ò¤¤¤¯¤Ä¤«¤ÎÉé²Ù¤Ç¼Â¹Ô¤·¤ÆÊÑÆ°¤¬Å¬Å٤˾®¤µ¤¤¤³¤È¤ò³Î¤«¤á¤ë¡£
+Âè 2 Ãʳ¬¤Ç¤Ï¡¢¶½Ì£¤ò»ý¤Ã¤¿¤¤¤¯¤Ä¤«¤ÎÉé²Ù¤Ç
+.B nhfsstone
+¤ò¼Â¹Ô¤·¡¢·ë²Ì¤òÊݸ¤¹¤ë¡£
+Âè 3 Ãʳ¬¤Ç¤Ï¡¢¥µ¡¼¥Ð¡¼¤ÎÀßÄê¤òÊѹ¹¤¹¤ë
+(¥á¥â¥ê¤òÁý¤ä¤¹¡¦¥Ç¥£¥¹¥¯¥³¥ó¥È¥í¡¼¥é¤òÊѹ¹¤¹¤ë¡¢¤Ê¤É)¡£
+ºÇ½ªÃʳ¬¤Ç¤Ï¡¢
+.B nhfsstone
+¤òƱ¤¸Éé²Ù¤ÇºÆÅټ¹Ԥ·¡¢·ë²Ì¤òÈæ³Ó¤¹¤ë¡£
+.\"O .SH SEE ALSO
+.SH ´ØÏ¢¹àÌÜ
+.LP
+.\"O The
+.\"O .B nhfsstone.c
+.\"O source file has comments that describe in detail the operation of
+.\"O of the program.
+¥½¡¼¥¹¥³¡¼¥É
+.B nhfsstone.c
+¤Ë¤Ï¡¢¥×¥í¥°¥é¥à¤Î¾ÜºÙ¤ÊÁàºî¤òµ½Ò¤·¤¿¥³¥á¥ó¥È¤¬½ñ¤«¤ì¤Æ¤¤¤ë¡£
+.\"O .SH ERROR MESSAGES
+.SH ¥¨¥é¡¼¥á¥Ã¥»¡¼¥¸
+.TP
+.B "illegal calls value"
+.\"O The
+.\"O .B calls
+.\"O argument following the
+.\"O .B \-c
+.\"O flag on the command line is not a positive number.
+¥³¥Þ¥ó¥É¥é¥¤¥ó¤Ç»ØÄꤵ¤ì¤¿
+.B \-c
+¥Õ¥é¥°¤Î¸å¤Î
+.B calls
+°ú¤¿ô¤¬Àµ¤Î¿ô¤Ç¤Ê¤¤¡£
+.TP
+.B "illegal load value"
+.\"O The
+.\"O .B load
+.\"O argument following the
+.\"O .B \-l
+.\"O flag on the command line is not a positive number.
+¥³¥Þ¥ó¥É¥é¥¤¥ó¤Ç»ØÄꤵ¤ì¤¿
+.B \-l
+¥Õ¥é¥°¤Î¸å¤Î
+.B load
+°ú¤¿ô¤¬Àµ¤Î¿ô¤Ç¤Ê¤¤¡£
+.TP
+.B "illegal time value"
+.\"O The
+.\"O .B time
+.\"O argument following the
+.\"O .B \-t
+.\"O flag on the command line is not a positive number.
+¥³¥Þ¥ó¥É¥é¥¤¥ó¤Ç»ØÄꤵ¤ì¤¿
+.B \-t
+¥Õ¥é¥°¤Î¸å¤Î
+.B time
+°ú¤¿ô¤¬Àµ¤Î¿ô¤Ç¤Ê¤¤¡£
+.TP
+.B "bad mix file"
+.\"O The
+.\"O .B mixfile
+.\"O file argument following the
+.\"O .B \-m
+.\"O flag on the command line could not be accessed.
+¥³¥Þ¥ó¥É¥é¥¤¥ó¤Ç»ØÄꤵ¤ì¤¿
+.B \-m
+¥Õ¥é¥°¤Î¸å¤Î
+.B mixfile
+¥Õ¥¡¥¤¥ë°ú¤¿ô¤Ë¥¢¥¯¥»¥¹¤Ç¤¤Ê¤¤¡£
+.TP
+.B "can't find current directory"
+.\"O The parent process couldn't find the pathname of the current directory.
+.\"O This usually indicates a permission problem.
+¿Æ¥×¥í¥»¥¹¤¬¥«¥ì¥ó¥È¥Ç¥£¥ì¥¯¥È¥ê¤Î¥Ñ¥¹Ì¾¤ò¸«¤Ä¤±¤ë¤³¤È¤¬¤Ç¤¤Ê¤¤¡£
+¸¢¸Â¤ÎÌäÂê¤òɽ¤·¤Æ¤¤¤ë¤³¤È¤¬Â¿¤¤¡£
+.TP
+.B "can't fork"
+.\"O The parent couldn't fork the child processes. This usually results from
+.\"O lack of resources, such as memory or swap space.
+¿Æ¥×¥í¥»¥¹¤¬»Ò¥×¥í¥»¥¹¤ò fork ¤Ç¤¤Ê¤¤¡£
+¥á¥â¥ê¤ä¥¹¥ï¥Ã¥×¶õ´Ö¤È¤¤¤Ã¤¿¥ê¥½¡¼¥¹¤ÎÉÔ¤ηë²Ì¤Ç¤¢¤ë¤³¤È¤¬Â¿¤¤¡£
+.TP
+.PD 0
+.B "can't open log file"
+.TP
+.B "can't stat log"
+.TP
+.B "can't truncate log"
+.TP
+.B "can't write sync file"
+.TP
+.B "can't write log"
+.TP
+.B "can't read log"
+.PD
+.\"O A problem occurred during the creation, truncation, reading or writing of the
+.\"O synchronization log file. The parent process creates the
+.\"O log file in /tmp and uses it to synchronize and communicate with its children.
+Ʊ´ü¥í¥°¥Õ¥¡¥¤¥ë¤ÎºîÀ®¡¦ÀÚ¤êµÍ¤á¡¦Æɤ߹þ¤ß¡¦½ñ¤½Ð¤·¤Î»þ¤ËÌäÂ꤬µ¯¤³¤Ã¤¿¡£
+¿Æ¥×¥í¥»¥¹¤¬¥í¥°¥Õ¥¡¥¤¥ë¤ò /tmp ¤ËºîÀ®¤·¡¢
+»Ò¥×¥í¥»¥¹¤È¤ÎƱ´ü¤äÄÌ¿®¤Ë»ÈÍѤ¹¤ë¡£
+.TP
+.PD 0
+.B "can't open test directory"
+.TP
+.B "can't create test directory"
+.TP
+.B "can't cd to test directory"
+.TP
+.B "wrong permissions on test dir"
+.TP
+.B "can't stat testfile"
+.TP
+.B "wrong permissions on testfile"
+.TP
+.B "can't create rename file"
+.TP
+.B "can't create subdir"
+.PD
+.\"O A child process had problems creating or checking the contents of its
+.\"O test directory. This is usually due to a permission problem (for example
+.\"O the test directory was created by a different user) or a full filesystem.
+»Ò¥×¥í¥»¥¹¤¬¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤ÎÆâÍƤòºîÀ®¡¦¸¡ºº¤¹¤ë¤È¤¤ËÌäÂ꤬µ¯¤³¤Ã¤¿¡£
+¸¢¸Â¤ÎÌäÂê (Î㤨¤Ð¡¢¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê¤¬
+Ê̤Υ桼¥¶¡¼¤Ë¤è¤Ã¤ÆºîÀ®¤µ¤ì¤Æ¤¤¤ë¾ì¹ç) ¤ä
+¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à¤¬¤¤¤Ã¤Ñ¤¤¤Ç¤¢¤ë¾ì¹ç¤¬Â¿¤¤¡£
+.TP
+.PD 0
+.B "bad mix format: unexpected EOF after 'nfs:'"
+.TP
+.B "bad mix format: can't find 'calls' value"
+.TP
+.B "bad mix format: unexpected EOF after 'calls'"
+.TP
+.B "bad mix format: can't find %d op values"
+.TP
+.B "bad mix format: unexpected EOF"
+.PD
+.\"O A problem occurred while parsing the
+.\"O .B mix
+.\"O file. The expected format of the file is the same as the output of
+.\"O the
+.\"O .BR nfsstat (8C)
+.\"O command when run with the "\-s" option.
+.B mix
+¥Õ¥¡¥¤¥ë¤ò²òÀϤ·¤Æ¤¤¤ë¤È¤¤ËÌäÂ꤬µ¯¤³¤Ã¤¿¡£
+¤³¤Î¥Õ¥¡¥¤¥ë¤ÎÁÛÄꤵ¤ì¤ë¥Õ¥©¡¼¥Þ¥Ã¥È¤Ï¡¢
+.BR nfsstat (8C)
+¥³¥Þ¥ó¥É¤ò "\-s" ¥ª¥×¥·¥ç¥ó¤Ç¼Â¹Ô¤·¤¿¤È¤¤Î½ÐÎϤÈƱ¤¸¤Ç¤¢¤ë¡£
+.TP
+.B "op failed: "
+.\"O One of the internal pseudo\-NFS operations failed. The name of the operation,
+.\"O e.g. read, write, lookup, will be printed along with an indication of the
+.\"O nature of the failure.
+ÆâÉôŪ¤Êµ¿»÷ NFS Áàºî¤Î 1 ¤Ä¤¬¼ºÇÔ¤·¤¿¡£
+Áàºî¤Î̾Á°¡¢Î㤨¤Ð read, write, lookup ¤¬¡¢¼ºÇԤμïÎà¤È¶¦¤Ëɽ¼¨¤µ¤ì¤ë¡£
+.TP
+.B "select failed"
+.\"O The select system call returned an unexpected error.
+select ¥·¥¹¥Æ¥à¥³¡¼¥ë¤¬Í½´ü¤»¤Ì¥¨¥é¡¼¤òÊÖ¤·¤¿¡£
+.\"O .SH BUGS
+.SH ¥Ð¥°
+.LP
+.\"O Running
+.\"O .B nhfsstone
+.\"O on a non\-NFS filesystem can cause the program to run forever because it
+.\"O uses the kernel NFS statistics to determine when enough calls have been made.
+.B nhfsstone
+¤ò NFS ¥Õ¥¡¥¤¥ë¥·¥¹¥Æ¥à°Ê³°¤Ç¼Â¹Ô¤¹¤ë¤È¡¢¥×¥í¥°¥é¥à¤¬±Êµ×¤ËÆ°ºî¤·Â³¤±¤ë¡£
+¤³¤Î¥×¥í¥°¥é¥à¤¬¥«¡¼¥Í¥ë¤Î NFS Åý·×¤ò»È¤Ã¤Æ¡¢
+½¼Ê¬¤Ê¸Æ¤Ó½Ð¤·²ó¿ô¤¬¼Â¹Ô¤µ¤ì¤¿¤«¤ò·èÄꤷ¤Æ¤¤¤ë¤¿¤á¤Ç¤¢¤ë¡£
+.LP
+.\"O .B Nhfsstone
+.\"O uses many file descriptors. The kernel on the client may
+.\"O have to be reconfigured to increase the number of available file table entries.
+.B nhfsstone
+¤¬Â¿¤¯¤Î¥Õ¥¡¥¤¥ë¥Ç¥£¥¹¥¯¥ê¥×¥¿¤ò»È¤¦¡£
+¥¯¥é¥¤¥¢¥ó¥È¤Î¥«¡¼¥Í¥ë¤òºÆÀßÄꤷ¡¢
+ÍøÍѲÄǽ¤Ê¥Õ¥¡¥¤¥ë¥Æ¡¼¥Ö¥ë¥¨¥ó¥È¥ê¤òÁý¤ä¤¹É¬Íפ¬¤¢¤ë¤«¤â¤·¤ì¤Ê¤¤¡£
+.LP
+.\"O Shell scripts that used
+.\"O .B nhfsstone
+.\"O will have to catch and ignore SIGUSR1 (see
+.\"O .BR signal (3)).
+.B nhfsstone
+¤ò»È¤¦¥·¥§¥ë¥¹¥¯¥ê¥×¥È¤Ï SIGUSR1
+.RB ( signal (3)
+¤ò»²¾È) ¤ò¥¥ã¥Ã¥Á¤·¤Æ̵»ë¤¹¤ëɬÍפ¬¤¢¤ë¡£
+.\"O This signal is
+.\"O used to synchronize the test processes. If the signal is not caught
+.\"O the shell that is running the script will be killed.
+¤³¤Î¥·¥°¥Ê¥ë¤Ï¥Æ¥¹¥È¥×¥í¥»¥¹¤òƱ´ü¤¹¤ë¤Î¤Ë»È¤ï¤ì¤ë¡£
+¤³¤Î¥×¥í¥»¥¹¤ò¥¥ã¥Ã¥Á¤·¤Ê¤¤¤È¡¢
+¥¹¥¯¥ê¥×¥È¤ò¼Â¹Ô¤·¤Æ¤¤¤ë¥·¥§¥ë¤¬ kill ¤µ¤ì¤ë¡£
+.\"O .SH FILES
+.SH ¥Õ¥¡¥¤¥ë
+.PD 0
+.TP 20
+.B /vmunix
+.\"O system namelist
+¥·¥¹¥Æ¥à¤Î¥Í¡¼¥à¥ê¥¹¥È
+.TP
+.B /dev/kmem
+.\"O kernel virtual memory
+¥«¡¼¥Í¥ë¤Î²¾ÁÛ¥á¥â¥ê
+.TP
+.B ./testdir*
+.\"O per process test directory
+¥×¥í¥»¥¹Ëè¤Î¥Æ¥¹¥È¥Ç¥£¥ì¥¯¥È¥ê
+.TP
+.B /tmp/nhfsstone%d
+.\"O process synchronization log file
+¥×¥í¥»¥¹¤ÎƱ´ü¥í¥°¥Õ¥¡¥¤¥ë
+.PD
OSDN Git Service
