On 2019-03-01 19:36, Jonathan Gibbons wrote:
On 03/01/2019 05:32 AM, Magnus Ihse Bursie wrote:
I think we should really get rid of sjavac since the relevant
benefits are already present in the default build, with the javac
server and the dependency plugin. The only possible benefit of
sjavac today would be more fine grained incremental build support,
but I doubt it works very well given that it's not being maintained.
Agree. I opened https://bugs.openjdk.java.net/browse/JDK-8219973.
/Magnus
Magnus, et al,
Be careful. There is as yet no "javac server". The server mechanism
is currently only available within sjavac.
There is a desire/goal to provide a "javac server". When first
suggested, it was considered blocked by the need of some
platform-specific features in the main `java.util.Process` API.
However, when the core-libs team looked at the RFE, it was not clear
that the work was actually required. The issue is the ability to
create a process that can outlive the parent; at one point, it was
believed that this was not easy/possible with existing Java API on all
necessary platforms (i.e all platforms supported by the build.)
Jon,
Your warning is noted. What I meant was that the build system currently
has two ways of providing javac server speedups, one of which works well
and is on by default, and one which has not been tested for a long time.
We should remove that option from the build system. I did not intend to
go into the finer details of what "really" consitutes sjavac.
/Magnus
-- Jon
