FW: RFR(XXXS): 8205996: JDK-8205906 broke the build on Mac
markus.gronlund at oracle.com
Thu Jun 28 12:09:00 UTC 2018
Erik and Tobias,
Thank you very much for your quick reviews.
You are right, I was stressing a bit getting a quick fix in place.
I will rework this section along the lines you suggest.
From: Erik Helin
Sent: den 28 juni 2018 13:47
To: Markus Gronlund <markus.gronlund at oracle.com>; hotspot-dev developers <hotspot-dev at openjdk.java.net>
Subject: Re: FW: RFR(XXXS): 8205996: JDK-8205906 broke the build on Mac
On 06/28/2018 01:23 PM, Markus Gronlund wrote:
> Widening this a bit for some quicker reviews (since this is a P1).
Given that the build is broken, I think you better just push this.
The patch can (and should) be improved IMO, please use strlen instead of "24" and "25". Two of the changed places are asserts, so the performance of strlen doesn't matter. The remaining place is part of parse_start_flight_recording_option which I would presume is only called when a jcmd is executed (or at JVM startup), so it can't be a hot path (please correct me if I'm wrong).
> Thanks in advance
> -----Original Message-----
> From: Markus Gronlund
> Sent: den 28 juni 2018 13:07
> To: hotspot-jfr-dev at openjdk.java.net; Tobias Hartmann
> <tobias.hartmann at oracle.com>
> Subject: RFR(XXXS): 8205996: JDK-8205906 broke the build on Mac
> Looks like the check-in for https://bugs.openjdk.java.net/browse/JDK-8205906 broke the product build when running on Mac (builds fine using the "debug" target that I tested before pushing.
> There is a P1 filed now to restore the Mach builds, so I would appreciate a quick ok.
> Webrev: http://cr.openjdk.java.net/~mgronlun/8205996/webrev00/
> Bug: https://bugs.openjdk.java.net/browse/JDK-8205996
> Summary: inlining char constants and lengths.
> Currently running a verification job in MACH5 (product build)
More information about the hotspot-dev