From owner-man-jp-reviewer@jp.freebsd.org  Sun Nov  7 23:36:47 1999
Received: (from daemon@localhost)
	by castle.jp.freebsd.org (8.9.3+3.2W/8.7.3) id XAA55680;
	Sun, 7 Nov 1999 23:36:47 +0900 (JST)
	(envelope-from owner-man-jp-reviewer@jp.FreeBSD.org)
Received: from light.imasy.or.jp (root@light.imasy.or.jp [202.227.24.4])
	by castle.jp.freebsd.org (8.9.3+3.2W/8.7.3) with ESMTP id XAA55612
	for <man-jp-reviewer@jp.freebsd.org>; Sun, 7 Nov 1999 23:36:39 +0900 (JST)
	(envelope-from mistral@imasy.or.jp)
Received: from tasogare.imasy.or.jp (mistral@tasogare.imasy.or.jp [202.227.24.5])
	by light.imasy.or.jp (8.9.3+3.2W/3.7W-light) with ESMTP id XAA08467
	for <man-jp-reviewer@jp.freebsd.org>; Sun, 7 Nov 1999 23:36:36 +0900 (JST)
	(envelope-from mistral@imasy.or.jp)
Received: from mistral.wind.prv (isdnb56.imasy.or.jp [202.227.24.184])
	by tasogare.imasy.or.jp (8.9.3+3.2W/3.7W-tasogare/smtpfeed 1.01) with ESMTP id XAA24752
	for <man-jp-reviewer@jp.freebsd.org>; Sun, 7 Nov 1999 23:36:32 +0900 (JST)
	(envelope-from mistral@imasy.or.jp)
Message-Id: <199911071436.XAA24752@tasogare.imasy.or.jp>
Received: (from yohta@localhost)
	by mistral.wind.prv (8.9.3/3.7Wpl2-990626) id XAA18685;
	Sun, 7 Nov 1999 23:35:36 +0900 (JST)
Date: Sun, 7 Nov 1999 23:35:36 +0900 (JST)
From: mistral@imasy.or.jp (Yoshihiko SARUMARU)
To: man-jp-reviewer@jp.freebsd.org
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-2022-JP
X-Mailer: mnews [version 1.21PL5] 1999-04/04(Sun)
Reply-To: man-jp-reviewer@jp.freebsd.org
Precedence: list
X-Distribute: distribute version 2.1 (Alpha) patchlevel 24e+990727
X-Sequence: man-jp-reviewer 1660
Subject: [man-jp-reviewer 1660] gsc.4
Errors-To: owner-man-jp-reviewer@jp.freebsd.org
Sender: owner-man-jp-reviewer@jp.freebsd.org
X-Originator: mistral@imasy.or.jp

  $B1n4]$G$9!#(B
  gsc.4 $B$G$9!#$J$s$+$h$/$o$+$i$J$$J8$,B?$+$C$?$G$9!D!D!#(B

> It is
> even questionable if external programs would not do this job better
> thereby not counting to the size of the kernel. 
$B$C$F$I$&$$$&0UL#$J$s$G$7$g$&!#(B
> $B$b$730It%W%m%0%i%`$,$3$N:n6H$r$h$j$&$^$/$G$-$J$+$C$?$H$7$F!"(B
> $B$=$N7k2L%+!<%M%k$NBg$-$5$,8:$k$+$I$&$+$9$i5?Ld$K;W$C$F$$$^$9!#(B
$B$HL5M}LpM}Lu$7$F$_$^$7$?$,!"$J$K$+0c$&$h$&$J5$$,$7$^$9!#(B

  $B$h$m$7$/$*4j$$$7$^$9!#(B

--- gsc.4~	Thu Nov  4 22:03:43 1999
+++ gsc.4	Sun Nov  7 23:29:46 1999
@@ -34,251 +34,241 @@
 .Dd January 9, 1995
 .Os FreeBSD
 .Dt GSC 4 i386
-.Sh NAME
+.Sh $BL>>N(B
 .Nm gsc
-.Nd a device driver for a handy scanner
-.Sh SYNOPSIS
+.Nd $B%O%s%G%#%9%-%c%JMQ%G%P%$%9%I%i%$%P(B
+.Sh $B=q<0(B
 .Fd #include <machine/gsc.h>
 .Pp
-.Cd Minor number bits: Ar uu d g p ...
+.Cd $B%^%$%JHV9f%S%C%H(B: Ar uu d g p ...
 .Bl -tag -width "uu " -compact -offset indent
 .It Ar uu
-unit gsc0 .. gsc3
+$B%f%K%C%H(B gsc0 .. gsc3
 .It Ar d
-selects logging of
-.Em debug
-messages
+.Em $B%G%P%C%0(B
+$B%a%C%;!<%8$N5-O?$rA*Br(B
 .It Ar g
-selects
-.Em bitmap
-vs.
-.Em graymap
-output
+.Em $B%S%C%H%^%C%W(B
+$B$^$?$O(B
+.Em $B%0%l!<%^%C%W(B
+$B=PNO$rA*Br(B
 .It Ar p
-selects
 .Em raw
-vs. portable
+$B$^$?$O(B portable
 .Em pnm
-output
+$B=PNO$rA*Br(B
 .El
 .Pp
-.Sh DESCRIPTION
-The
+.Sh $B2r@b(B
 .Nm gsc
-character device driver currently handles only the
-Genius GS-4500 handy scanner. It operates in pure dma modus, although
-the hardware could be set up to work with irq. I had neither enough
-documentation nor experience in writing interrupt driven device
-drivers.
-.Pp
-The device can operate at four different 
-.Em resolutions :
+$B%-%c%i%/%?%G%P%$%9%I%i%$%P$O!":#$N$H$3$m(B Genius GS-4500 $B%O%s%G%#%9%-%c%J(B
+$B$N$_$r07$C$F$$$^$9!#$3$N%I%i%$%P$O%O!<%I%&%'%"$,(B irq $B$r;HMQ$7$F(B
+$BF0:n2DG=$G$"$C$F$b!"40A4$K(B dma $BJ}<0$G@)8f$r9T$J$$$^$9!#(B
+$B;d$K$O3d$j9~$_6nF07?$N%G%P%$%9%I%i%$%P$r=q$/$N$K==J,$J(B
+$B%I%-%e%a%s%H$b7P83$b$"$j$^$;$s!#(B
+.Pp
+$B%G%P%$%9$O(B 4 $B$D$N0[$J$C$?(B
+.Em $B2rA|EY(B
+$B$K@)8f$G$-$^$9(B:
 100, 200,
-300 and 400dpi. It produces a simple bitmap with the most significant
-bit at the left side. The driver can optionally output the famous and
-likely simple portable bitmap file format
+300, 400dpi$B!#%G%P%$%9$O:G>e0L$N%S%C%H$,:8$K$/$k%7%s%W%k$J%S%C%H%^%C%W$r(B
+$B@8@.$7$^$9!#%I%i%$%P$O(B Jef Poskanzer $B$K$h$k!"(B
+$B$h$/CN$i$l$F$$$FE,EY$K%7%s%W%k$J%]!<%?%V%k%S%C%H%^%C%W%U%!%$%k7A<0(B
 .Xr pbm 5
-by Jef Poskanzer.
-Thus the scans can easily processed by any graphic package around
-(
+$B$G=PNO$9$k$3$H$b$G$-$^$9!#(B
+$B$7$?$,$C$F%9%-%c%s$N7k2L$O$=$NJU$K$"$k$I$N$h$&$J%0%i%U%#%C%/%Q%C%1!<%8(B
+($B0lIt$NL>A0$r5s$2$k$J$i(B
 .Xr xpaint 1 ,
 .Xr xv 1 ,
 .Xr xli 1
-only to name some of them ...).  In 
+$B$J$I(B) $B$G4JC1$K=hM}$9$k$3$H$,$G$-$^$9!#(B
 .Em raw
-mode a
-bit which is set means a black pixel because the scanner detects black
-points on white paper. On the other hand, because pnm format describes
-intensities of electron beams in video screens a set bit in 
+$B%b!<%I$G$O!"%9%-%c%J$OGr$$;f$N>e$N9u$$E@$r8!=P$9$k$?$a!"(B
+$B%;%C%H$5$l$?%S%C%H$O9u$$%T%/%;%k$r0UL#$7$^$9!#(B
+$B5U$K!"(Bpnm $B%U%)!<%^%C%H$O%S%G%*2hLL$NEE;R%S!<%`$N8wEY$r5-=R$9$k$?$a!"(B
 .Em pbm
-mode means a white pixel.
+$B%b!<%I$K$*$$$F$O!"%;%C%H$5$l$?%S%C%H$OGr$$%T%/%;%k$r0UL#$7$^$9!#(B
 .Pp
-The 
-.Em width
-of the output bitmap is fixed as given by the
-resolution value. However, the 
-.Em height
-of the bitmap must be
-supplied in 
+$B=PNO$5$l$?%S%C%H%^%C%W$N(B
+.Em $BI}(B
+$B$OM?$($i$l$?2rA|EYCM$K$h$C$F8GDj$5$l$^$9!#$7$+$7$J$,$i(B
 .Em pnm
-mode since the driver must know at what time the
-'end-of-file' shall be reached. With this feature you are able to
-directly copy the scanner output into a pbm file with
-.Xr cat .
-Of course you can obtain a similar effect by using
-.Xr dd 1
-with the driver in 
+$B%b!<%I$G$O!"%I%i%$%P$O$$$D(B '$B%U%!%$%k$N=*$o$j(B' $B$,8!=P$5$l$?$+$rCN$C$F$$$k(B
+$B$O$:$J$N$G!"%S%C%H%^%C%W$N(B
+.Em $B9b$5(B
+$B$O<+F0$G@_Dj$5$l$^$9!#(B
+$B$3$N5!G=$K$h$j!"(B
+.Xr cat
+$B$rMQ$$$F%9%-%c%J=PNO$r(B pbm $B%U%!%$%k$KD>@\%3%T!<$9$k$3$H$,$G$-$^$9!#(B
+$B$b$A$m$s(B
+$B%I%i%$%P$r(B
 .Em raw
-mode.
+$B%b!<%I$G;HMQ$7(B
+.Xr dd 1
+$B$rMQ$$$F;w$?8z2L$rF@$k$3$H$O$G$-$^$9!#(B
 .Pp
-The 
-.Em graymap
-output mode is not yet implemented into the driver.
-It is even questionable if external programs would not do this job
-better thereby not counting to the size of the kernel. Even though, I
-do not know of tools which produce a graymap from a halftone bitmap.
+.Em $B%0%l!<%^%C%W(B
+$B=PNO%b!<%I$O$^$@%I%i%$%P$K<BAu$5$l$F$$$^$;$s!#(B
+$B$b$730It%W%m%0%i%`$,$3$N:n6H$r$h$j$&$^$/$G$-$J$+$C$?$H$7$F!"(B
+$B$=$N7k2L%+!<%M%k$NBg$-$5$,8:$k$+$I$&$+$9$i5?Ld$K;W$C$F$$$^$9!#(B
+$B$?$@$?$H$($=$&$G$"$C$?$H$7$F$b!"Cf4VD4%S%C%H%^%C%W$+$i%0%l!<%^%C%W$r(B
+$B@8@.$9$k%D!<%k$r;d$OCN$j$^$;$s!#(B
 .Pp
-The ioctl requests that are served by
 .Nm gsc
-are listed below.
-There is a utility, called sgsc(1), that provides access to these
-requests from within shell.
+$B$,=hM}$9$k(B ioctl $B%j%/%(%9%H$O0J2<$KNs5s$5$l$F$$$^$9!#(B
+sgsc(1) $B$H8F$P$l$k%f!<%F%#%j%F%#$r;H$&$H!"%7%'%k$+$i$3$l$i$N(B
+$B%j%/%(%9%H$K%"%/%;%9$G$-$^$9!#(B
 .Bl -tag -width Ds
 .It GSC_SRES int
-Set the
-.Em resolution
-value. If this call is made after the first
-read access to the device there will be no effect unless the device is
-closed and opened again.
+
+.Em $B2rA|EY(B
+$B$NCM$r@_Dj$7$^$9!#%G%P%$%9$X$N:G=i$NFI$_9~$_%"%/%;%9$N8e$K!"(B
+$B$3$N8F$S=P$7$r9T$J$C$?>l9g$O!"%G%P%$%9$r%/%m!<%:$7$F:F$S%*!<%W%s$7$J$$8B$j(B
+$B8z2L$r;}$A$^$;$s!#(B
 .It GSC_GRES int
-Get current resolution in dots per inch (dpi).
+$B8=:_$N2rA|EY$r(B 1 $B%$%s%A$"$?$j$N%I%C%H?t(B (dpi) $B$G<hF@$7$^$9!#(B
 .It GSC_SRESSSW void
-Set resolution value from selector switch. The driver must be in an
-open though untouched state otherwise the request will fail and
+$B2rA|EY$NCM$r%;%l%/%?%9%$%C%A$G@_Dj$7$^$9!#%I%i%$%P$O%*!<%W%s$5$l$F$$$k$,(B
+$B?($l$i$l$F$$$J$$>uBV$G$"$kI,MW$,$"$j$^$9!#$=$&$G$J$$>l9g$O%j%/%(%9%H$O(B
+$B<:GT$7!"(B
 .Xr errno 2
-is set to EBUSY.
+$B$K$O(B EBUSY $B$,F~$j$^$9!#(B
 .It GSC_SWIDTH int
-Set the 
-.Em width
-of the bitmap. Actually, this is an alternative
-way of setting the resolution, since any allowed resolution matches
-exactly one width. Allowed are listed in the table below.
+$B%S%C%H%^%C%W$N(B
+.Em $BI}(B
+$B$r@_Dj$7$^$9!#$I$N$h$&$J@_Dj2DG=$J2rA|EY$b$^$5$7$/0l$D$NI}$KBP1~$9$k$N$G!"(B
+$B<B:]$O$3$N%j%/%(%9%H$O2rA|EY$r@_Dj$9$kBe$o$j$N<jCJ$H$J$j$^$9!#(B
+$B@_Dj2DG=$JCM$O2<$NI=$KNs5s$5$l$F$$$^$9!#(B
 .Bl -tag -width resolution -compact -offset indent
-.It resolution
-width
+.It $B2rA|EY(B
+$BI}(B
 .It 100 dpi
-424 pixels
+424 $B%T%/%;%k(B
 .It 200 dpi
-840 pixels
+840 $B%T%/%;%k(B
 .It 300 dpi
-1264 pixels
+1264 $B%T%/%;%k(B
 .It 400 dpi
-1648 pixels
+1648 $B%T%/%;%k(B
 .It ?
-1696 pixels
+1696 $B%T%/%;%k(B
 .It ?
-2544 pixels
+2544 $B%T%/%;%k(B
 .It ?
-3648 pixels
+3648 $B%T%/%;%k(B
 .El
 .Pp
-Values which are not reported in the above table will cause the ioctl
-call to fail with 
+$B>e$NI=$K=q$+$l$F$$$J$$CM$O(B ioctl $B8F$S=P$7$K<:GT$7!"(B
 .Xr errno 2
-set to
-.Er EINVAL .
-.Pp
-As you can see, there are width values > 1696. This does, however, not
-mean that you can obtain scanned lines longer than the width of your
-scanner or by higher resolutions. Actually, the resolution is selected
-by only by the hardware switch. Any line that is longer than what is
-defined for the actual resolution will be undefined (usually white) on
-the right part that is exceeding the standard line.
+$B$K$O(B
+.Er EINVAL
+$B$,F~$j$^$9!#(B
+.Pp
+$B8+$?$H$*$j!"(B1696 $B$h$j$bBg$-$JI}$NCM$,$"$j$^$9!#$7$+$7!"$3$N$3$H$O(B
+$B%9%-%c%J$NI}$h$j$bD9$$9T$r%9%-%c%s$G$-$?$j!"(B
+$B$b$C$H9b$$2rA|EY$G%9%-%c%s$G$-$?$j$9$k$3$H$r0UL#$9$k$o$1$G$O$"$j$^$;$s!#(B
+$B<B:]!"2rA|EY$O%O!<%I%&%'%"%9%$%C%A$N$_$K$h$C$FA*Br$5$l$^$9!#(B
+$B<B:]$N2rA|EY$GDj5A$5$l$F$$$k0J>e$ND9$5$N9T$G!"I8=`$N9T$ND9$5$r1[$($?(B
+$B1&ItJ,$ODj5A$5$l$F$$$^$;$s(B ($BDL>o$OGr$K$J$j$^$9(B)$B!#(B
 .It GSC_GWIDTH int
-Get current width of the bitmap in pixels.
+$B8=:_$N%S%C%H%^%C%W$NI}$r%T%/%;%k?t$G<hF@$7$^$9!#(B
 .It GSC_SHEIGHT int
-Set the 
-.Em height
-of the bitmap in 
 .Em pnm
-mode. This is actually
-a limit on the amount of lines scannable after the first read
-operation. When the limit is reached read will return 0. However, the
-device is turned off only when a close is performed (either
-explicitly or implicitly on exit of the calling process).
+$B%b!<%I$G$N%S%C%H%^%C%W$N(B
+.Em $B9b$5(B
+$B$r@_Dj$7$^$9!#(B
+$B<B:]$K$O$3$N%j%/%(%9%H$O:G=i$NFI$_9~$_A`:n$N8e!"%9%-%c%s$G$-$k9T?t$N9g7W$r(B
+$B@)8B$7$^$9!#@)8BCM$KC#$7$?;~!"FI$_9~$_$O(B 0 $B$rJV$7$^$9!#(B
+$B$7$+$7!"(B($BL@<(E*$"$k$$$O0EL[E*$K8F$S=P$7%W%m%;%9$N=*N;;~$K(B)
+$B%G%P%$%9$,%/%m!<%:$5$l$?;~$K$N$_!"EE8;$O@ZCG$5$l$^$9!#(B
 .It GSC_GHEIGHT int
-Get the current height of the bitmap.
+$B8=:_$N%S%C%H%^%C%W$N9b$5$r<hF@$7$^$9!#(B
 .It GSC_SBLEN int
-Set the length of the buffer used internally to do the DMA transfer.
-The buffer length is supplied in lines of the bitmap. Since the buffer
-size limit is (currently) 0x3000 bytes the maximum number of lines
-allowed will vary with the width of each line. This upper limit is
-checked before it overwrites the current value and pases an ENOMEM in
-the
+DMA $BE>Aw$KFbIt$G;HMQ$9$k%P%C%U%!$ND9$5$r@_Dj$7$^$9!#(B
+$B%P%C%U%!$ND9$5$O%S%C%H%^%C%W$N9T?t$GM?$($i$l$^$9!#(B
+$B%P%C%U%!!<$N%5%$%:$O(B ($B8=:_(B) 0x3000 $B%P%$%H$K@)8B$5$l$F$$$k$N$G!"(B
+$B5v$5$l$k%i%$%s$N:GBg?t$O3F9T$NI}$K$h$C$FJQ2=$7$^$9!#(B
+$B$3$N>e8B$O8=:_$NCM$r>e=q$-$7(B
 .Xr errno 2 
-variable. However, since the bitmap width can change
-after a buffer length was selected a read request may fail with ENOMEM
-if the buffer length turns out too high. It is generally wise to
-choose long buffers rather than go save in order to obtain better
-output.
+$BJQ?t(B $B$K(B ENOMEM $B$rF~$l$F$7$^$&A0$K%A%'%C%/$5$l$^$9!#(B
+$B$7$+$7!"%P%C%U%!9T?t$rA*Br$7$?8e$b%S%C%H%^%C%W$NI}$rJQ99$9$k$3$H$O(B
+$B$G$-$k$N$G!"%P%C%U%!9T?t$,Bg$-$9$.$k$3$H$,$o$+$C$?;~$OFI$_9~$_%j%/%(%9%H$O(B
+ENOMEM $B$G<:GT$9$k$+$b$7$l$^$;$s!#0lHLE*$K$O$h$jNI$$=PNO$rF@$k$?$a$K$O!"(B
+$B@aLs$9$k$h$j$bD9$$%P%C%U%!$rA*Br$9$kJ}$,8-$$$G$9!#(B
 .It GSC_GBLEN int
-Get the current buffer length in lines.
+$B8=:_$N%P%C%U%!$ND9$5$r9T?t$G<hF@$7$^$9!#(B
 .It GSC_SBTIME int
-Set the timeout for the completion of reading one buffer. Since a
-handy scanner is a human/computer interface timeout values are usually
-higher than those of a flat scanner. Default is 15 seconds. After
-timeout is reached the read operation will fail with EBUSY. Note that
-the timeout timer starts anew for each buffer to be read and thus does
-not cause you to scan faster for longer images. BLEN/BTIME is similar
-as MIN/TIME in termios(4).
+1 $B$D$N%P%C%U%!$rFI$_9~$_=*$o$k$^$G$N%?%$%`%"%&%H;~4V$r@_Dj$7$^$9!#(B
+$B%O%s%G%#%9%-%c%J$O%U%i%C%H%9%-%c%J$h$j$b!"?M4V(B / $B%3%s%T%e!<%?(B
+$B%$%s%?%U%'!<%9$N%?%$%`%"%&%H;~4V$,D9$$$N$G!"%G%U%)%k%H$G$O(B 15 $BIC$K(B
+$B$J$C$F$$$^$9!#%?%$%`%"%&%H;~4V8e$O!"FI$_9~$_A`:n$O(B EBUSY $B$G<:GT$7$^$9!#(B
+$B%?%$%`%"%&%H!&%?%$%^$O$=$l$>$l$N%P%C%U%!$,FI$_9~$^$l$kEY$K(B
+$B?7$?$K3+;O$5$l$k$N$G!"D9$$%$%a!<%8$rAa$/%9%-%c%s$G$-$k$h$&$K$J$k(B
+$B$o$1$G$O$J$$$3$H$KCm0U$7$F$/$@$5$$!#(BBLEN/BTIME $B$O(B termios(4) $B$N(B
+MIN/TIME $B$H;w$F$$$^$9!#(B
 .It GSC_GBTIME int
-Get the current buffer timeout.
+$B8=:_$N%P%C%U%!$N%?%$%`%"%&%H;~4V$r<hF@$7$^$9!#(B
 .El
 .Pp
-All ioctl requests that modify a parameter except GSC_SBTIME do not
-have an effect on an ongoing scan process, i.e. after the first read
-request that follows open. You must close the device and open it again
-for the new selections to take effect. Consequently, the selections
-are not reset when you close or open the device.
-.Pp
-Similarly, requests that read a value do not report the value that is
-used for the ongoing scan process. The values needed during the scan
-process are saved when it starts and thus are not accessed by ioctl
-requests.
-.Pp
-The BTIME value does, however, have an immediate effect on the ongoing
-scan. Thus the timeout can for example be set to long until the user
-starts scanning.  It can then be set to a short amount to react
-(nearly) immediately when the user stops. Note that the user should be
-left time to at least fill one buffer without having to haste.
+$B%Q%i%a!<%?$rJQ99$9$k(B GSC_SBTIME $B0J30$N$9$Y$F$N(B ioctl $B%j%/%(%9%H$O!"(B
+$B8=:_?J9TCf$N%9%-%c%s%W%m%;%9$K$O1F6A$7$^$;$s!#$9$J$o$A:G=i$NFI$_9~$_(B
+$B%j%/%(%9%H;~$N@_Dj$,%G%P%$%9$N%*!<%W%sCf$OB3$-$^$9!#1F6A$r5Z$\$9$K$O(B
+$B%G%P%$%9$r%/%m!<%:$7$F%*!<%W%s$7D>$9I,MW$,$"$j$^$9!#$7$?$,$C$F!"(B
+$B@_Dj$O%G%P%$%9$r%/%m!<%:$9$k$+%*!<%W%s$9$k$^$G%j%;%C%H$5$l$^$;$s!#(B
+.Pp
+$BF1MM$K!"@_DjCM$rFI$`%j%/%(%9%H$O8=:_?J9TCf$N%9%-%c%s%W%m%;%9$G(B
+$B;H$o$l$F$$$kCM$rJs9p$7$^$;$s!#%9%-%c%s%W%m%;%9$,I,MW$H$9$kCM$O(B
+$B3+;O;~$KJ]B8$5$l$F$$$k$N$G!"(Bioctl $B%j%/%(%9%H$G$O%"%/%;%9$G$-$^$;$s!#(B
+.Pp
+$B$7$+$7(B BTIME $BCM$O8=:_?J9TCf$N%9%-%c%s$K$9$0$KH?1G$5$l$^$9!#(B
+$B$7$?$,$C$F%?%$%`%"%&%H;~4V$O!"Nc$($P%f!<%6$,%9%-%c%s$r3+;O$9$k$^$G$N(B
+$BD9$$;~4V$K@_Dj$9$k$3$H$,$G$-$^$9!#(B
+$B$^$?%f!<%6$,Dd;_$7$?$i(B ($B$[$\(B) $B$9$0$KH?1~$G$-$k$h$&$JC;$$;~4V$K(B
+$B@_Dj$9$k$3$H$b$G$-$^$9!#(B
+$B%f!<%6$,92$F$k$3$H$J$/0l$D$N%P%C%U%!$rKd$a$i$l$k$@$1$N;~4V$O!":GDc$G$b;D$9(B
+$BI,MW$,$"$k$3$H$KCm0U$7$F$/$@$5$$!#(B
 .Pp
-Note that the 
 .Em pbm
-versus 
+$B$H(B
 .Em raw
-mode selection is done by the
-minor number not by ioctl requests. In 
+$B$N%b!<%IA*Br$O!"(Bioctl $B%j%/%(%9%H$G$O$J$/!"%^%$%JHV9f$K$h$C$F9T$J$o$l$k(B
+$B$3$H$KCm0U$7$F$/$@$5$$!#(B
 .Em raw
-mode the selected
-height of the bitmap will have no effect.
-.Sh FILES
+$B%b!<%I$G$O!"A*Br$5$l$?%S%C%H%^%C%W$N9b$5$O$J$s$N8z2L$b;}$A$^$;$s!#(B
+.Sh $B4XO"%U%!%$%k(B
 .Bl -tag -width /dev/gsc0pd
 .It Pa /dev/gsc0
-device node for 
 .Em raw
-output, has minor number 0.
+$B=PNO$N$?$a$N%G%P%$%9%N!<%I$G!"%^%$%JHV9f$O(B 0 $B$G$9!#(B
 .It Pa /dev/gsc0d
-device node for 
 .Em raw
-output emitting 
-.Em debug
-messages if the
-GSCDEBUG option was given at compile time, has minor number 32.
+$B=PNO$N$?$a$N%G%P%$%9%N!<%I$G!"(B
+$B%3%s%Q%$%k;~$K(B GSCDEBUG $B%*%W%7%g%s$,@_Dj$5$l$F$$$?$i!"(B
+.Em $B%G%P%C%/(B
+$B%a%C%;!<%8$r=PNO$7$^$9!#%^%$%JHV9f$O(B 32 $B$G$9!#(B
 .It Pa /dev/gsc0p
-device node for output in 
 .Em pbm
-file format, has minor number 8.
+$B%U%!%$%k%U%)!<%^%C%H=PNO$N$?$a$N%G%P%$%9%N!<%I$G!"%^%$%JHV9f$O(B 8 $B$G$9!#(B
 .It Pa /dev/gsc0pd
 device node for 
 .Em pbm
-and 
+$B$H(B
 .Em debug
-mode, has minor number 40.
+$B%b!<%I$N$?$a$N%G%P%$%9%N!<%I$G!"%^%$%JHV9f$O(B 40 $B$G$9!#(B
 .PB
-.Sh EXAMPLES
+.Sh $B;HMQNc(B
 .Bd -literal -offset indent
 dd if=/dev/gsc0 of=rawfile bs=(width/8) count=(height)
-
 cat /dev/gsc0p > pbmfile
 .Ed
-.Sh DIAGNOSTICS
+.Sh $B?GCG(B
 .Bl -tag
 .It GSCDEBUG
-When you define this name as an `option' in the kernel configuration
-you can get debug output if you access the driver with a minor number
-whose debug bit (i.e. bit 5 out of 7) is set.
+$B%+!<%M%k@_Dj$G$3$NL>>N$r(B `option' $B$H$7$FDj5A$9$k$H!"(B
+$B%G%P%C%0%S%C%H(B ($B$9$J$o$A(B 7 $B%S%C%HCf$N%S%C%H(B 5) $B$,%;%C%H$5$l$?%^%$%JHV9f(B
+$B$G%I%i%$%P$K%"%/%;%9$7$?>l9g%G%P%C%0=PNO$rF@$k$3$H$,$G$-$^$9!#(B
 .El
-.Sh SEE ALSO
+.Sh $B4XO"9`L\(B
 .Xr cat 1 ,
 .Xr dd 1 ,
 .Xr pbmtopgm 1 ,
@@ -291,11 +281,11 @@
 .Xr read 2 ,
 .Xr termios 4 ,
 .Xr pbm 5
-.Sh AUTHORS
+.Sh $B:n<T(B
 .An Gunther Schadow Aq gusw@fub46.zedat.fu-berlin.de
-.Sh BUGS
-Even though the scanner device has a little switch by which you should
-be able to select one of the four resolution modes, I could not yet
-determine how to read its status. Unless this is not fixed the driver
-depends on the value passed by means of ioctl(2) which need not match
-what is selected by the hardware.
+.Sh $B%P%0(B
+$B$?$H$(%9%-%c%J%G%P%$%9$K!"(B4 $B$D$N2rA|EY$N$&$A(B 1 $B$D$rA*Br$G$-$k$h$&$K(B
+$B$J$C$F$$$k>.$5$J%9%$%C%A$,$"$C$?$H$7$F$b!";d$K$O$^$@$=$N>uBV$rCN$kJ}K!$,(B
+$B$o$+$C$F$$$^$;$s!#(B
+$B$3$NLdBj$,=$@5$5$l$J$$8B$j!"%O!<%I%&%'%"$K$h$C$FA*Br$5$l$?$b$N$H0lCW$9$k(B
+$BI,MW$N$J$$(B ioctl(2) $B$K$h$C$F$d$j$H$j$5$l$kCM$K!"%I%i%$%P$O0MB8$7$^$9!#(B




--
$B1n4](B($B5l@+(B: $BB@ED(B)$BK'I'(B (Yoshihiko SARUMARU)
E-mail: mistral@imasy.or.jp	URL: http://www.imasy.or.jp/~mistral/
