<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="/rss.xsl.xml"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
<channel>
    <title>Changes in memcached.h</title>
    <description></description>
    <language>en</language>
    <copyright>Copyright 2015</copyright>
    <generator>Java</generator><item>
        <title>ee461d11 - slabs reassigns works with chunks and chunked items.</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#ee461d11</link>
        <description>slabs reassigns works with chunks and chunked items.also fixes the new LRU algorithm to balance by total bytes used rather thantotal chunks used, since total chunks used isn&apos;t tracked for multi-chunkitems.also fixes a bug where the lru limit wasn&apos;t being utilized for HOT_LRUalso some cleanup from previous commits.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 12 Jul 2016 02:06:56 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>b05653f9 - chunked item second checkpoint</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#b05653f9</link>
        <description>chunked item second checkpointcan actually fetch items now, and fixed a few bugs with storage/freeing.added fetching for binprot.added some basic tests.many tests still fail for various reasons, and append/prepend isn&apos;t fixed yet.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Fri, 01 Jul 2016 21:41:29 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>0567967a - chunked items checkpoint commit</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#0567967a</link>
        <description>chunked items checkpoint commitcan set and store large items via asciiprot. gets/append/prepend/binprot notimplemented yet.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Thu, 30 Jun 2016 21:02:47 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>68c64594 - use X macros to remove several struct iterations</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#68c64594</link>
        <description>use X macros to remove several struct iterationsAdding new stats would require updating in too many locations. No longer haveto. Could extend this a bit to do the actual stats printing, but for clarity&apos;ssake I&apos;ll decide on that later.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 28 Jun 2016 00:08:08 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>cb01d504 - clean up global stats code a little.</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#cb01d504</link>
        <description>clean up global stats code a little.tons of stats were left in the global stats structure that&apos;re no longer used,and it looks like we kept accidentally adding new ones in there. There&apos;s alsoan unused mutex.Split global stats into `stats` and `stats_state`. initialize via memset,reset only `stats` via memset, removing several places where stats values getrepeated. Looks much cleaner and should be less error prone.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Mon, 27 Jun 2016 23:26:40 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>80ce0108 - add get_flushed counter, fix expired_unfetched</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#80ce0108</link>
        <description>add get_flushed counter, fix expired_unfetchedget_flushed ticks when directly fetching something that classified as expired.LRU crawler was ticking expired_unfetched for unfetched _flushed_ values.should no longer do that.also removes get_expired from the global stats list. I missed that on review.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Mon, 27 Jun 2016 19:31:13 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>d704f2c0 - add -o watcher_logbuf_size=N worker_logbuf_size=N</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#d704f2c0</link>
        <description>add -o watcher_logbuf_size=N worker_logbuf_size=Nthe defaults should just about always be used, but this is now tunable forcompleteness. Also handy for writing tests.This also hopefully makes the watcher.t test not flaky anymore.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Fri, 24 Jun 2016 05:20:03 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>83ba6bd9 - memcached idle connection killer</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#83ba6bd9</link>
        <description>memcached idle connection killer

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Wed, 03 Sep 2014 19:46:05 +0000</pubDate>
        <dc:creator>Jay Grizzard &lt;jgrizzard@box.com&gt;</dc:creator>
    </item>
<item>
        <title>c7fbcceb - logger endpoints for first release</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#c7fbcceb</link>
        <description>logger endpoints for first releaseswapping &quot;RAWCMDS&quot; for internal hooks on when items are fetched or stored.This doesn&apos;t map 1:1 with commands, ie: a store is internally a fetch thenstore, so two log lines are generated. An ascii multiget one make one log lineper key fetched.It&apos;s a good place to start. Need to come back and refactor parts of logger.cagain, then convert all prints in the codebase to log entries.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Thu, 16 Jun 2016 23:18:14 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>0503b5e2 - some global stats for bg logger</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#0503b5e2</link>
        <description>some global stats for bg loggerfour obvious ones. have a handy place to do the summarization when the loggerwakes up for its run, avoiding the locks.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Mon, 13 Jun 2016 04:57:02 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>916fff36 - initial logger code.</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#916fff36</link>
        <description>initial logger code.Logs are written to per-thread buffers. A new background thread aggregates thelogs, further processes them, then writes them to any &quot;watchers&quot;.Logs can have the time added to them, and all have a GID so they can be putback into strict order.This is an early preview. Code needs refactoring and a more complete set ofoptions. All watchers are also stuck viewing the global feed of logs, even ifthey asked for different data.As of this commit there&apos;s no way to toggle the &quot;stderr&quot; watcher.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Wed, 18 Nov 2015 09:49:53 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>9517c656 - bump some global stats to 64bit uints</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#9517c656</link>
        <description>bump some global stats to 64bit uintstotal_items is pretty easy to overflow. Upped some of the others just in case.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 31 May 2016 23:02:54 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>6895d23e - Implement get_expired stats</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#6895d23e</link>
        <description>Implement get_expired stats

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Wed, 09 Dec 2015 00:32:49 +0000</pubDate>
        <dc:creator>sergiocarlos &lt;carloschilazo@gmail.com&gt;</dc:creator>
    </item>
<item>
        <title>b1debc4c - try harder to save items</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#b1debc4c</link>
        <description>try harder to save itemspreviously the slab mover would evict items if the new chunk was within theslab page being moved. now it will do an inline reclaim of the chunk and tryuntil it runs out of memory.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Sat, 10 Oct 2015 03:59:04 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>8fa54f7e - split rebal_evictions into _nomem and _samepage</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#8fa54f7e</link>
        <description>split rebal_evictions into _nomem and _samepagegross oversight putting two conditions into the same variable. now can tell ifwe&apos;re evicting because we&apos;re hitting the bottom of the free memory pool, or ifwe keep trying to rescue items into the same page as the one being cleared.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Thu, 08 Oct 2015 07:28:27 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>6ee8daef - call STATS_LOCK() less in slab mover.</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#6ee8daef</link>
        <description>call STATS_LOCK() less in slab mover.uses the slab_rebal struct to summarize stats, more occasionally grabbing theglobal lock to fill them in, instead.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 06 Oct 2015 02:04:02 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>d6e96467 - first half of new slab automover</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#d6e96467</link>
        <description>first half of new slab automoverIf any slab classes have more than two pages worth of free chunks, attempt tofree one page back to a global pool.Create new concept of a slab page move destination of &quot;0&quot;, which is a globalpage pool. Pages can be re-assigned out of that pool during allocation.Combined with item rescuing from the previous patch, we can safely shufflepages back to the reassignment pool as chunks free up naturally. This shouldbe a safe default going forward. Users should be able to decide to free ormove pages based on eviction pressure as well. This is coming up in anothercommit.This also fixes a calculation of the NOEXP LRU size, and completely removesthe old slab automover thread. Slab automove decisions will now be part of thelru maintainer thread.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 29 Sep 2015 09:48:02 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>004e2211 - slab mover rescues valid items with free chunks</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#004e2211</link>
        <description>slab mover rescues valid items with free chunksDuring a slab page move items are typically ejected regardless of theirvalidity. Now, if an item is valid and free chunks are available in the sameslab class, copy the item over and replace it.It&apos;s up to external systems to try to ensure free chunks are available beforemoving a slab page. If there is no memory it will simply evict them as normal.Also adds counters so we can finally tell how often these cases happen.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Tue, 29 Sep 2015 00:32:49 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
<item>
        <title>a1f269ee - Record and report on time spent in listen_disabled</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#a1f269ee</link>
        <description>Record and report on time spent in listen_disabled

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Thu, 24 Jan 2013 14:08:20 +0000</pubDate>
        <dc:creator>Ian Miell &lt;ian.miell@gmail.com&gt;</dc:creator>
    </item>
<item>
        <title>369845f0 - fix major off by one issue</title>
        <link>http://172.16.0.5:8080/history/memcached-1.4.29/memcached.h#369845f0</link>
        <description>fix major off by one issuenone of my machines could repro a crash, but it&apos;s definitely wrong :/ Verysad.

            List of files:
            /memcached-1.4.29/memcached.h</description>
        <pubDate>Fri, 24 Apr 2015 07:03:49 +0000</pubDate>
        <dc:creator>dormando &lt;dormando@rydia.net&gt;</dc:creator>
    </item>
</channel>
</rss>
