RFR: JDK-8041487: Fix proper dependencies for correct incremental build of javadocs

Jonathan Gibbons jonathan.gibbons at oracle.com
Tue Apr 22 21:35:22 UTC 2014

The comments at the end of the webrev could be improved as well at some 
You can barely see the code for all the dust and spiders.  I think it 
talks about
"large 32 bits machines" but surely that can't be right... :-)

1200 #release version of core packages ########
1202 # The rel-coredocs and rel-docs targets were added by Eric Armstrong. rel-coredocs
1203 # assumes the kind of large, 32-bit machine used in the javapubs group's docs-release
1204 # process. It specifies memory settings accordingly to maximize performance.
1205 #
1206 # The performance settings, like the sanity check, are most important for the core
1207 # docs--the platform APIs. Running javadoc on those APIs takes a significant amount
1208 # of time and memory. Setting the initial heap size as large as possible is important
1209 # to prevent thrashing as the heap grows. Setting the maximum as large as necessary
1210 # is also important to keep the job from failing.
1211 #
1212 #    -J-Xmx512 sets a maximum of 512, which became necessary in 6.0
1213 #    -J-Xms256 sets starting size to 256 (default is 8)
1214 #
1215 # rel-coredocs also includes a sanity check to help ensure that BUILD_NUMBER and
1216 # MILESTONE are specified properly when docs are built outside of the normal release
1217 # engineering process, with the intention of releasing them on the web or in a downloaded

On 04/22/2014 08:04 AM, Erik Joelsson wrote:
> (adding javadoc-dev)
> On 2014-04-22 15:29, Erik Joelsson wrote:
>> (reposting with correct subject)
>> On 2014-04-22 15:28, Erik Joelsson wrote:
>>> I recently had to work with make/Javadoc.gmk and felt that it needed 
>>> some attention. This patch makes it behave correctly for incremental 
>>> builds and reduces the log output on the default log level to much 
>>> more manageable levels.
>>> I realize that we should probably rewrite this even more to reduce 
>>> all the code duplication, but choose to leave that for later.
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8041487
>>> Webrev: http://cr.openjdk.java.net/~erikj/8041487/webrev.root.01/
>>> /Erik

More information about the build-dev mailing list