CINXE.COM

<?xml version="1.0" encoding="UTF-8" standalone="yes"?> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>LKML: "Artem B. Bityuckiy": [RFC] CryptoAPI & Compression</title><link href="/css/message.css" rel="stylesheet" type="text/css" /><link href="/css/wrap.css" rel="alternate stylesheet" type="text/css" title="wrap" /><link href="/css/nowrap.css" rel="stylesheet" type="text/css" title="nowrap" /><link href="/favicon.ico" rel="shortcut icon" /><script src="/js/simple-calendar.js" type="text/javascript"></script><script src="/js/styleswitcher.js" type="text/javascript"></script><link rel="alternate" type="application/rss+xml" title="lkml.org : last 100 messages" href="/rss.php" /><link rel="alternate" type="application/rss+xml" title="lkml.org : last messages by "Artem B. Bityuckiy"" href="/groupie.php?aid=28528" /><script> var _paq = window._paq = window._paq || []; /* tracker methods like "setCustomDimension" should be called before "trackPageView" */ _paq.push(["setDoNotTrack", true]); _paq.push(["disableCookies"]); _paq.push(['trackPageView']); _paq.push(['enableLinkTracking']); (function() { var u="//m.lkml.org/"; _paq.push(['setTrackerUrl', u+'matomo.php']); _paq.push(['setSiteId', '1']); var d=document, g=d.createElement('script'), s=d.getElementsByTagName('script')[0]; g.async=true; g.src=u+'matomo.js'; s.parentNode.insertBefore(g,s); })(); </script></head><body onload="es.jasper.simpleCalendar.init();" itemscope="itemscope" itemtype="http://schema.org/BlogPosting"><table border="0" cellpadding="0" cellspacing="0"><tr><td width="180" align="center"><a href="/"><img style="border:0;width:135px;height:32px" src="/images/toprowlk.gif" alt="lkml.org" /></a></td><td width="32">聽</td><td class="nb"><div><a class="nb" href="/lkml"> [lkml]</a> 聽 <a class="nb" href="/lkml/2005"> [2005]</a> 聽 <a class="nb" href="/lkml/2005/3"> [Mar]</a> 聽 <a class="nb" href="/lkml/2005/3/25"> [25]</a> 聽 <a class="nb" href="/lkml/last100"> [last100]</a> 聽 <a href="/rss.php"><img src="/images/rss-or.gif" border="0" alt="RSS Feed" /></a></div><div>Views: <a href="#" class="nowrap" onclick="setActiveStyleSheet('wrap');return false;">[wrap]</a><a href="#" class="wrap" onclick="setActiveStyleSheet('nowrap');return false;">[no wrap]</a> 聽 <a class="nb" href="/lkml/mheaders/2005/3/25/104" onclick="this.href='/lkml/headers'+'/2005/3/25/104';">[headers]</a>聽 <a href="/lkml/bounce/2005/3/25/104">[forward]</a>聽 </div></td><td width="32">聽</td></tr><tr><td valign="top"><div class="es-jasper-simpleCalendar" baseurl="/lkml/"></div><div class="threadlist">Messages in this thread</div><ul class="threadlist"><li class="root"><a href="/lkml/2005/3/25/104">First message in thread</a></li><li class="origin"><a href="/lkml/2005/3/25/277">"Artem B. Bityuckiy"</a><ul><li><a href="/lkml/2005/3/25/277">Herbert Xu</a><ul><li><a href="/lkml/2005/3/26/34">"Artem B. Bityuckiy"</a></li><li><a href="/lkml/2005/3/28/118">"Artem B. Bityuckiy"</a><ul><li><a href="/lkml/2005/3/29/92">Herbert Xu</a><ul><li><a href="/lkml/2005/3/29/122">"Artem B. Bityuckiy"</a></li></ul></li><li><a href="/lkml/2005/3/31/74">Herbert Xu</a><ul><li><a href="/lkml/2005/4/3/58">"Artem B. Bityuckiy"</a></li></ul></li></ul></li></ul></li></ul></li></ul></td><td width="32" rowspan="2" class="c" valign="top"><img src="/images/icornerl.gif" width="32" height="32" alt="/" /></td><td class="c" rowspan="2" valign="top" style="padding-top: 1em"><table><tr><td><table><tr><td class="lp">Subject</td><td class="rp" itemprop="name">[RFC] CryptoAPI & Compression</td></tr><tr><td class="lp">From</td><td class="rp" itemprop="author">"Artem B. Bityuckiy" <></td></tr><tr><td class="lp">Date</td><td class="rp" itemprop="datePublished">Fri, 25 Mar 2005 19:08:20 +0300</td></tr></table></td><td></td></tr></table><pre itemprop="articleBody">Hello Herbert and others,<br /><br />I'm working on cleaning-up the JFFS3 compression stuff. JFFS3 contains a<br />number of compressors which actually shouldn't be there as they are<br />platform-independent and generic. So we want to move them to the generic<br />part of the Linux kernel instead of storing them in fs/jffs2/. And we<br />were going to use CryptoAPI to access the compressors.<br /><br />But I've hit on a problem - CryptoAPI's compression method isn't<br />flexible enough for us.<br /><br />CryptoAPI's compress method is:<br /><br />int crypto_compress(struct crypto_tfm *tfm,<br /> const u8 *src, unsigned int slen,<br /> u8 *dst, unsigned int *dlen);<br /><br />*src - input data;<br />slen - input data length;<br />*dst - output data;<br />*dlen - on input - output buffer length, on output - the length of the<br />compressed data;<br /><br />The crypto_compress() API call either compresses all the input data or<br />returns error.<br /><br />In JFFS2 we need something more flexible. Te following is what we want:<br /><br />int crypto_compress_ext(struct crypto_tfm *tfm,<br /> const u8 *src, unsigned int *slen,<br /> u8 *dst, unsigned int *dlen);<br /><br />*src - input data;<br />*slen - on input - input data length, on output - the amount of data<br />that were actually compressed.<br />*dst - output data;<br />*dlen - on input - output buffer length, on output - the length of the<br />compressed data;<br /><br />This would allow us (and we really need this) to provide a large input<br />data buffer, a small output data buffer and to ask the compressor to<br />compress as much input data as it can to fit (in the compressed form) to<br />the output buffer. To put it differently, we often have a large input,<br />and several smalloutput buffers, and we want to compress the input data<br />to them.<br /><br />I offer to extend the CryptoAPI and add an "extended compress" API call<br />with the above mentioned capabilities. We might as well just change the<br />crypto_compress() and all its users.<br /><br />Alternatively, we may create some kind of "Compression API" but I don't<br />like this idea...<br /><br />Comments?<br /><br />-- <br />Best Regards,<br />Artem B. Bityuckiy,<br />St.-Petersburg, Russia.<br /><br />-<br />To unsubscribe from this list: send the line "unsubscribe linux-kernel" in<br />the body of a message to majordomo@vger.kernel.org<br />More majordomo info at <a href="http://vger.kernel.org/majordomo-info.html">http://vger.kernel.org/majordomo-info.html</a><br />Please read the FAQ at <a href="http://www.tux.org/lkml/">http://www.tux.org/lkml/</a><br /><br /></pre></td><td width="32" rowspan="2" class="c" valign="top"><img src="/images/icornerr.gif" width="32" height="32" alt="\" /></td></tr><tr><td align="right" valign="bottom"> 聽 </td></tr><tr><td align="right" valign="bottom">聽</td><td class="c" valign="bottom" style="padding-bottom: 0px"><img src="/images/bcornerl.gif" width="32" height="32" alt="\" /></td><td class="c">聽</td><td class="c" valign="bottom" style="padding-bottom: 0px"><img src="/images/bcornerr.gif" width="32" height="32" alt="/" /></td></tr><tr><td align="right" valign="top" colspan="2"> 聽 </td><td class="lm">Last update: 2005-04-06 13:30 聽聽 [from the cache]<br />漏2003-2020 <a href="http://blog.jasper.es/"><span itemprop="editor">Jasper Spaans</span></a>|hosted at <a href="https://www.digitalocean.com/?refcode=9a8e99d24cf9">Digital Ocean</a> and my Meterkast|<a href="http://blog.jasper.es/categories.html#lkml-ref">Read the blog</a></td><td>聽</td></tr></table><script language="javascript" src="/js/styleswitcher.js" type="text/javascript"></script></body></html>