Re: [9] JDK-8184306: zlib 1.2.11 upgrade triggers j.u.zip.Deflater regression

2017-07-17 Thread Alan Bateman
On 13/07/2017 18:08, Martin Buchholz wrote: : - configure on linux reports: --with-zlib use zlib from build system or OpenJDK source (system, bundled) [bundled] Does this need updating to "system" ? Yes, except when on Windows, but that can be updated

Re: [9] JDK-8184306: zlib 1.2.11 upgrade triggers j.u.zip.Deflater regression

2017-07-13 Thread Brian Burkhalter
On Jul 12, 2017, at 11:52 PM, Alan Bateman wrote: > On 13/07/2017 00:06, Xueming Shen wrote: >> : >> >> My reading of the 1.2.11 changes suggests the root cause is that the internal >> state "deflate_state.high_water" is not being reset correctly/appropriately >> (to 0?) >> in deflateReset/defl

Re: [9] JDK-8184306: zlib 1.2.11 upgrade triggers j.u.zip.Deflater regression

2017-07-13 Thread Martin Buchholz
Random drive-by comments. - consider adding a link to the JDK bug to the zlib bug. - configure on linux reports: --with-zlib use zlib from build system or OpenJDK source (system, bundled) [bundled] Does this need updating to "system" ?

Re: [9] JDK-8184306: zlib 1.2.11 upgrade triggers j.u.zip.Deflater regression

2017-07-12 Thread Alan Bateman
On 13/07/2017 00:06, Xueming Shen wrote: : My reading of the 1.2.11 changes suggests the root cause is that the internal state "deflate_state.high_water" is not being reset correctly/appropriately (to 0?) in deflateReset/deflateResetKeep/lm_init(). AND the change of one of the conditions in

[9] JDK-8184306: zlib 1.2.11 upgrade triggers j.u.zip.Deflater regression

2017-07-12 Thread Xueming Shen
Hi, Please help review the change for #8184306. issue: https://bugs.openjdk.java.net/browse/JDK-8184306 webrev: http://cr.openjdk.java.net/~sherman/8184306/ Background: It appears the basic compression functionality of j.u.Deflater is broken when (1) the deflater's compression level/strate