Miscellaneous jar improvements

[Tim Bell]

Andrew wrote:

>>>> changeset:   1283:ff32c270102a
>>>> user:        martin
>>>> date:        Mon Jun 22 21:07:20 2009 -0700
>>>> summary:     6853806: Prefer (cd $dir && jar) to jar -C for performance reasons
> 
> Weird, I can't find that changeset.  Got a URL?

Here is is in the JDK7 master forest:
 http://hg.openjdk.java.net/jdk7/jdk7/jdk/rev/ff32c270102a


Martin wrote:

>>>> All told, there were at least 3 separate changesets in openjdk7 to
>>>> make building rt.jar faster.

More went by recently, including:

6834805: Improve jar -C performance
 http://hg.openjdk.java.net/jdk7/jdk7/jdk/rev/ce55eb6668d9

6854795: Miscellaneous improvements to "jar"
 http://hg.openjdk.java.net/jdk7/jdk7/jdk/rev/0cabe1192c8b

Discussed here:
 http://mail.openjdk.java.net/pipermail/core-libs-dev/2009-July/001993.html

>>> But you put the lot into http://hg.openjdk.java.net/jdk6/jdk6/jdk/rev/b35f1e5075a4,
>>> right?  That's what we have in IcedTea 6.
>> No, I was too lazy to backport 6853806.http://hg.openjdk.java.net/jdk7/jdk7/jdk/raw-rev/0cabe1192c8b
>> (Perhaps I determined it was unnecessary?  I don't recall)
>>
>> The changes to the jar command benefit all users of the jar command,
>> unlike 6853806, which only benefits us poor folks building openjdk.
> 
> Aha!  Proof, if any more were needed, that it would be better to have
> had the backport discussion online... ;-)
> 
> Out of interest, why does changing "jar -C" to (cd $dir && jar) make
> anything go faster?

> contributed by Jeremy Manson, who writes:
>
> """Basically, rt.jar seems to be built by taking a huge long list of
> directories:
> jar -C dir class -C dir class -C dir class <repeat 16000 times>
> All of these directories are *exactly the same place*.  However, the jar command
> gives each a separate entry in the big list of directories.
> For each class, it iterates through all 16000 identical entries to find the
> right directory.  Using HashSet reduces the size of this set to 1."""

See the full thread here:
http://mail.openjdk.java.net/pipermail/core-libs-dev/2009-April/001516.html


HTH-
Tim