00000064.htm

一份很好的linux入门资料

Both&nbsp;of&nbsp;these&nbsp;defaults&nbsp;can&nbsp;be&nbsp;overridden&nbsp;by&nbsp;command-line&nbsp;switches&nbsp;in&nbsp;this&nbsp;&nbsp;<BR>
format:&nbsp;patch-kernel&nbsp;[&nbsp;sourcedir&nbsp;[
&nbsp;<BR>
patchdir&nbsp;]&nbsp;].&nbsp;Patch-kernel&nbsp;will&nbsp;abort&nbsp;if&nbsp;any&nbsp;part&nbsp;of&nbsp;the&nbsp;patch&nbsp;fails,&nbsp;but&nbsp;if&nbsp;&nbsp;<BR>
the&nbsp;patch&nbsp;applies&nbsp;cleanly&nbsp;it&nbsp;will&nbsp;invoke&nbsp;find,&nbsp;which
&nbsp;<BR>
will&nbsp;delete&nbsp;all&nbsp;of&nbsp;the&nbsp;*.orig&nbsp;files&nbsp;which&nbsp;patch&nbsp;leaves&nbsp;behind.&nbsp;
&nbsp;<BR>

&nbsp;<BR>
If&nbsp;you&nbsp;prefer&nbsp;to&nbsp;see&nbsp;the&nbsp;output&nbsp;of&nbsp;commands,&nbsp;or&nbsp;perhaps&nbsp;you&nbsp;would&nbsp;rather&nbsp;keep&nbsp;&nbsp;<BR>
the&nbsp;*.orig&nbsp;files&nbsp;until&nbsp;you&nbsp;are&nbsp;certain&nbsp;the&nbsp;patched
&nbsp;<BR>
source&nbsp;compiles,&nbsp;running&nbsp;patch&nbsp;directly&nbsp;(with&nbsp;the&nbsp;patch&nbsp;located&nbsp;in&nbsp;the&nbsp;kernel&nbsp;&nbsp;<BR>
source&nbsp;top-level&nbsp;directory,&nbsp;as&nbsp;outlined&nbsp;above)&nbsp;has
&nbsp;<BR>
been&nbsp;very&nbsp;reliable&nbsp;in&nbsp;my&nbsp;experience.&nbsp;In&nbsp;order&nbsp;to&nbsp;avoid&nbsp;uncompressing&nbsp;the&nbsp;&nbsp;<BR>
patch&nbsp;before&nbsp;applying&nbsp;it&nbsp;a&nbsp;simple&nbsp;pipe&nbsp;will&nbsp;do&nbsp;the&nbsp;trick:&nbsp;
&nbsp;<BR>

&nbsp;<BR>
gzip&nbsp;-cd&nbsp;patchXX.gz&nbsp;|&nbsp;patch&nbsp;-p1
&nbsp;<BR>

&nbsp;<BR>
or:&nbsp;
&nbsp;<BR>

&nbsp;<BR>
bzip2&nbsp;-dc&nbsp;patchXX.bz2&nbsp;|&nbsp;patch&nbsp;-p1&nbsp;
&nbsp;<BR>

&nbsp;<BR>
After&nbsp;the&nbsp;patch&nbsp;has&nbsp;been&nbsp;applied&nbsp;the&nbsp;find&nbsp;utility&nbsp;can&nbsp;be&nbsp;used&nbsp;to&nbsp;check&nbsp;for&nbsp;&nbsp;<BR>
rejected&nbsp;files:&nbsp;
&nbsp;<BR>

&nbsp;<BR>
find&nbsp;.&nbsp;-name&nbsp;\*.rej&nbsp;
&nbsp;<BR>

&nbsp;<BR>
At&nbsp;first&nbsp;the&nbsp;syntax&nbsp;of&nbsp;this&nbsp;command&nbsp;is&nbsp;confusing.&nbsp;The&nbsp;period&nbsp;indicates&nbsp;that&nbsp;&nbsp;<BR>
find&nbsp;should&nbsp;look&nbsp;in&nbsp;the&nbsp;current&nbsp;directory&nbsp;and&nbsp;recursively
&nbsp;<BR>
in&nbsp;all&nbsp;subdirectories&nbsp;beneath&nbsp;it.&nbsp;Remember&nbsp;the&nbsp;period&nbsp;should&nbsp;have&nbsp;a&nbsp;space&nbsp;&nbsp;<BR>
both&nbsp;before&nbsp;and&nbsp;after&nbsp;it.&nbsp;The&nbsp;backslash&nbsp;before&nbsp;the
&nbsp;<BR>
wildcard&nbsp;&quot;*&quot;&nbsp;&quot;escapes&quot;&nbsp;the&nbsp;asterisk&nbsp;in&nbsp;order&nbsp;to&nbsp;avoid&nbsp;confusing&nbsp;the&nbsp;shell,&nbsp;&nbsp;<BR>
for&nbsp;which&nbsp;an&nbsp;asterisk&nbsp;has&nbsp;another&nbsp;meaning.&nbsp;If&nbsp;find&nbsp;locates
&nbsp;<BR>
any&nbsp;*.rej&nbsp;files&nbsp;it&nbsp;will&nbsp;print&nbsp;the&nbsp;filenames&nbsp;on&nbsp;the&nbsp;screen.&nbsp;If&nbsp;find&nbsp;exits&nbsp;&nbsp;<BR>
without&nbsp;any&nbsp;visible&nbsp;output&nbsp;it's&nbsp;nearly&nbsp;certain&nbsp;the&nbsp;patch
&nbsp;<BR>
applied&nbsp;correctly.&nbsp;
&nbsp;<BR>

&nbsp;<BR>
Another&nbsp;job&nbsp;for&nbsp;find&nbsp;is&nbsp;to&nbsp;remove&nbsp;the&nbsp;*.orig&nbsp;files:&nbsp;
&nbsp;<BR>

&nbsp;<BR>
find&nbsp;.&nbsp;-name&nbsp;\*.orig&nbsp;-print0&nbsp;|&nbsp;xargs&nbsp;-0r&nbsp;rm&nbsp;-f&nbsp;
&nbsp;<BR>

&nbsp;<BR>
This&nbsp;command&nbsp;is&nbsp;sufficiently&nbsp;cumbersome&nbsp;to&nbsp;type&nbsp;that&nbsp;it&nbsp;would&nbsp;be&nbsp;a&nbsp;good&nbsp;&nbsp;<BR>
candidate&nbsp;for&nbsp;a&nbsp;new&nbsp;shell&nbsp;alias.&nbsp;A&nbsp;line&nbsp;in&nbsp;your&nbsp;~/.bashrc
&nbsp;<BR>
file&nbsp;such&nbsp;as:&nbsp;
&nbsp;<BR>

&nbsp;<BR>
alias&nbsp;findorig&nbsp;'find&nbsp;.&nbsp;-name&nbsp;\*.orig&nbsp;-print0&nbsp;|&nbsp;xargs&nbsp;-0r&nbsp;rm&nbsp;-f'&nbsp;
&nbsp;<BR>

&nbsp;<BR>
will&nbsp;allow&nbsp;just&nbsp;typing&nbsp;findorig&nbsp;to&nbsp;invoke&nbsp;the&nbsp;above&nbsp;command.&nbsp;The&nbsp;single&nbsp;&nbsp;<BR>
quotes&nbsp;in&nbsp;the&nbsp;alias&nbsp;definition&nbsp;are&nbsp;necessary&nbsp;if&nbsp;an
&nbsp;<BR>
aliased&nbsp;command&nbsp;contains&nbsp;spaces.&nbsp;In&nbsp;order&nbsp;to&nbsp;use&nbsp;a&nbsp;new&nbsp;alias&nbsp;without&nbsp;logging&nbsp;&nbsp;<BR>
out&nbsp;and&nbsp;then&nbsp;back&nbsp;in&nbsp;again,&nbsp;just&nbsp;type&nbsp;source
&nbsp;<BR>
~/.bashrc&nbsp;at&nbsp;the&nbsp;prompt.&nbsp;
&nbsp;<BR>

&nbsp;<BR>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Incidental&nbsp;Comments&nbsp;and&nbsp;Conclusion
&nbsp;<BR>

&nbsp;<BR>
While&nbsp;putting&nbsp;this&nbsp;article&nbsp;together&nbsp;I&nbsp;upgraded&nbsp;the&nbsp;version&nbsp;of&nbsp;patch&nbsp;on&nbsp;my&nbsp;&nbsp;<BR>
machine&nbsp;from&nbsp;version&nbsp;2.1&nbsp;to&nbsp;version&nbsp;2.5.&nbsp;Both&nbsp;of&nbsp;these
&nbsp;<BR>
versions&nbsp;come&nbsp;from&nbsp;the&nbsp;current&nbsp;FSF/GNU&nbsp;maintainers.&nbsp;Immediately&nbsp;I&nbsp;noticed&nbsp;&nbsp;<BR>
that&nbsp;the&nbsp;default&nbsp;output&nbsp;of&nbsp;version&nbsp;2.5&nbsp;has&nbsp;been
&nbsp;<BR>
changed,&nbsp;with&nbsp;less&nbsp;information&nbsp;appearing&nbsp;on&nbsp;the&nbsp;screen.&nbsp;Gone&nbsp;is&nbsp;Larry&nbsp;Wall's&nbsp;&nbsp;<BR>
&quot;...hmm&quot;&nbsp;which&nbsp;used&nbsp;to&nbsp;appear&nbsp;while&nbsp;patch&nbsp;was
&nbsp;<BR>
attempting&nbsp;to&nbsp;determine&nbsp;the&nbsp;proper&nbsp;lines&nbsp;to&nbsp;patch.&nbsp;The&nbsp;output&nbsp;of&nbsp;version&nbsp;2.5&nbsp;&nbsp;<BR>
is&nbsp;simply&nbsp;a&nbsp;list&nbsp;of&nbsp;messages&nbsp;such&nbsp;as&nbsp;&quot;patching&nbsp;file
&nbsp;<BR>
[filename]&quot;,&nbsp;rather&nbsp;than&nbsp;the&nbsp;more&nbsp;copious&nbsp;information&nbsp;shown&nbsp;by&nbsp;earlier&nbsp;&nbsp;<BR>
versions.&nbsp;Admittedly,&nbsp;the&nbsp;information&nbsp;scrolled&nbsp;by&nbsp;too&nbsp;quickly
&nbsp;<BR>
to&nbsp;read,&nbsp;but&nbsp;the&nbsp;output&nbsp;could&nbsp;be&nbsp;redirected&nbsp;to&nbsp;a&nbsp;file&nbsp;for&nbsp;later&nbsp;perusal.&nbsp;This&nbsp;&nbsp;<BR>
change&nbsp;doesn't&nbsp;affect&nbsp;the&nbsp;functionality&nbsp;of&nbsp;the&nbsp;program,
&nbsp;<BR>
but&nbsp;does&nbsp;lessen&nbsp;the&nbsp;human&nbsp;element.&nbsp;It&nbsp;seems&nbsp;to&nbsp;me&nbsp;that&nbsp;touches&nbsp;such&nbsp;as&nbsp;the&nbsp;&nbsp;<BR>
old&nbsp;&quot;...hmm&quot;&nbsp;messages,&nbsp;as&nbsp;well&nbsp;as&nbsp;comments&nbsp;in
&nbsp;<BR>
source&nbsp;code,&nbsp;are&nbsp;valuable&nbsp;in&nbsp;that&nbsp;they&nbsp;remind&nbsp;the&nbsp;user&nbsp;that&nbsp;a&nbsp;program&nbsp;is&nbsp;the&nbsp;&nbsp;<BR>
result&nbsp;of&nbsp;work&nbsp;performed&nbsp;by&nbsp;a&nbsp;living,&nbsp;breathing&nbsp;human
&nbsp;<BR>
being,&nbsp;rather&nbsp;than&nbsp;a&nbsp;sterile&nbsp;collection&nbsp;of&nbsp;bits.&nbsp;The&nbsp;old&nbsp;behavior&nbsp;can&nbsp;be&nbsp;&nbsp;<BR>
restored&nbsp;by&nbsp;appending&nbsp;the&nbsp;switch&nbsp;--verbose&nbsp;to&nbsp;the&nbsp;patch
&nbsp;<BR>
command-line,&nbsp;but&nbsp;I'm&nbsp;sure&nbsp;that&nbsp;many&nbsp;users&nbsp;either&nbsp;won't&nbsp;be&nbsp;aware&nbsp;of&nbsp;the&nbsp;&nbsp;<BR>
option&nbsp;or&nbsp;won't&nbsp;bother&nbsp;to&nbsp;type&nbsp;it&nbsp;in.&nbsp;Another&nbsp;difference
&nbsp;<BR>
between&nbsp;2.1&nbsp;and&nbsp;2.5&nbsp;is&nbsp;that&nbsp;the&nbsp;*.orig&nbsp;back-up&nbsp;files&nbsp;aren't&nbsp;created&nbsp;unless&nbsp;&nbsp;<BR>
patch&nbsp;is&nbsp;given&nbsp;the&nbsp;-b&nbsp;option.&nbsp;
&nbsp;<BR>

&nbsp;<BR>
Patch&nbsp;is&nbsp;not&nbsp;strictly&nbsp;necessary&nbsp;for&nbsp;an&nbsp;end-user&nbsp;who&nbsp;isn't&nbsp;interested&nbsp;in&nbsp;&nbsp;<BR>
trying&nbsp;out&nbsp;and&nbsp;providing&nbsp;bug-reports&nbsp;for&nbsp;&quot;bleeding-edge&quot;
&nbsp;<BR>
software&nbsp;and&nbsp;kernels,&nbsp;but&nbsp;often&nbsp;the&nbsp;most&nbsp;interesting&nbsp;developments&nbsp;in&nbsp;the&nbsp;&nbsp;<BR>
Linux&nbsp;world&nbsp;belong&nbsp;in&nbsp;this&nbsp;category.&nbsp;It&nbsp;isn't&nbsp;difficult&nbsp;to&nbsp;get
&nbsp;<BR>
the&nbsp;hang&nbsp;of&nbsp;using&nbsp;patch,&nbsp;and&nbsp;the&nbsp;effort&nbsp;will&nbsp;be&nbsp;amply&nbsp;repaid.&nbsp;
&nbsp;<BR>

&nbsp;<BR>

&nbsp;<BR>
Last&nbsp;modified:&nbsp;Mon&nbsp;31&nbsp;Aug&nbsp;1998&nbsp;
&nbsp;<BR>

&nbsp;<BR>

&nbsp;<BR>

&nbsp;<BR>
&nbsp;<BR>
--&nbsp;<BR>
&nbsp;&nbsp;&nbsp;&nbsp;白马带著她一步步的回到中原。白马已经老了，只能慢慢的走，&nbsp;<BR>
但终是能回到中原的。江南有杨柳、桃花，有燕子、金鱼……&nbsp;<BR>
汉人中有的是英俊勇武的少年，倜傥潇洒的少年……但这个美&nbsp;<BR>
丽的姑娘就像古高昌国人那样固执:&nbsp;<BR>
&nbsp;<BR>
&nbsp;&nbsp;&nbsp;&nbsp;「那都是很好很好的，可是我偏不喜欢。」&nbsp;&nbsp;<BR>
&nbsp;<BR>
※&nbsp;来源:·BBS&nbsp;水木清华站&nbsp;bbs.net.tsinghua.edu.cn·[FROM:&nbsp;202.99.18.67]&nbsp;&nbsp;<BR>


<CENTER><H1>BBS水木清华站∶精华区</H1></CENTER>



</BODY>

</HTML>