Hi Serguei, I wasn’t aware that the jdk/com/sun/jdi tests would test libjdwp, I just found hotspot/jtreg/serviceability/jdwp when searching for jdwp in the tests. But thanks to your hints I know better now. In fact I ran the jdi tests when testing my change for 8193183.
I made a new webrev which includes the removal of inStream_endOfInput: http://cr.openjdk.java.net/~clanger/webrevs/8193258.1/. Can you please approve this. This time I ran the jdi tests without issues and also did builds on Windows, linux x86, AIX, Solaris and Mac. ☺ Best regards Christoph From: [email protected] [mailto:[email protected]] Sent: Samstag, 9. Dezember 2017 06:27 To: Langer, Christoph <[email protected]>; [email protected]; Chris Plummer <[email protected]> Subject: Re: RFR (S): 8193258: Better usage of JDWP HEADER SIZE Hi Christoph, You need to run at least the jdk/com/sun/jdi tests. Thanks, Serguei On 12/8/17 13:07, Langer, Christoph wrote: Hi Serguei, I did only run hotspot/jtreg/serviceability/jdwp, didn’t find a lot more in that area. I’m hoping/waiting for Chris’ tests then. I agree, I will then remove inStream_endOfInput. If something like that is needed for future developments, it can easily be added again. Best regards Christoph From: [email protected]<mailto:[email protected]> [mailto:[email protected]] Sent: Freitag, 8. Dezember 2017 21:12 To: Langer, Christoph <[email protected]><mailto:[email protected]>; [email protected]<mailto:[email protected]>; Chris Plummer <[email protected]><mailto:[email protected]> Subject: Re: RFR (S): 8193258: Better usage of JDWP HEADER SIZE Hi Christoph, The fix looks good to me. What tests did you run? On 12/8/17 07:07, Langer, Christoph wrote: Hi, Here’s a proposal to clean up the usage of the JDWP header size within the source code of libjdwp. Bug: https://bugs.openjdk.java.net/browse/JDK-8193258 WebRev: http://cr.openjdk.java.net/~clanger/webrevs/8193258.0/<http://cr.openjdk.java.net/%7Eclanger/webrevs/8193258.0/> As for inStream.c, I’m wondering wether inStream_endOfInput shall be removed? It seems to be unused… I'm inclined to remove it. Otherwise, it must be: 417 return (stream->left <= 0); Thanks, Serguei Best regards Christoph
