----- Original Message ----- > On 09/13/2013 10:28 PM, Andrew Hughes wrote: > > ----- Original Message ----- > >> Please review three following patches those fix jdk6 builds on > >> windows-i586. These patches were also contained in a patchset proposed > >> earlier for icedtea6 [1]. > >> > >> All changes were tested on windows 2000 with Microsoft Visual Studio > >> .NET 2003 that is minimum build environment [2] and the only working > >> compiler for windows-i586. > >> > >> 1. Hotspot SIZE_MAX patch, webrev [3], webrev download [4]. > >> SIZE_MAX macro definition, copied from [5]. _WIN32 checks are necessary > >> because MSVS2003 doesn't have stdint.h > >> > > > > This one from before looks fine to me. I'm just not sure about whether we > > want to include stdint.h on later VS versions. Would they ever be used? > Checked it, neither VS2003 (used for i586) nor Platform SDK April 2005 > (used for amd64) has stdint.h. So I've removed it. > Updated webrev - > http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_sizemax.01/ >
No, it shouldn't be removed. This file is also used by non-Windows platforms. I was thinking in the opposite direction; that some Windows platforms may have finally added this header. > > > > This is all alien to me :) > > > >> 2. Hotspot variadic macro patch, webrev [6], webrev download [7] > >> Turn empty variadic macro into simple one, MSVS2003 doesn't support > >> varmacros and this one is used only with two arguments anyway. Also in > >> jdk7u this varmacro was later removed [8]. > >> > > > > Again, this looks fine, though needs an #ifdef as this is shared by UNIX > > code. > Added _WIN32 check and tested build on windows and linux. > Updated webrev - > http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_varmacro.01/ > Thanks. I want to keep the non-Windows build unchanged if possible. > > > >> 3. JDK function parameter rename, webrev [9], webrev download [10] > >> It seems parameter has been accidentally renamed on backport of this > >> commit [11] > > > > I have a feeling this was more a rewrite than a backport, so no wonder the > > untested Windows side got broken. > > > >> > >> PS: I've signed OCA some time ago - "alexkasko" in signatories list [12]. > >> > >> > >> [1] > >> http://mail.openjdk.java.net/pipermail/distro-pkg-dev/2013-July/024037.html > >> [2] > >> http://hg.openjdk.java.net/jdk6/jdk6/raw-file/tip/README-builds.html#MBE > >> [3] http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_sizemax.00/ > >> [4] http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_sizemax.00.zip > >> [5] > >> http://hg.openjdk.java.net/jdk7u/jdk7u-dev/jdk/file/84101d9f6c3a/src/share/native/common/sizecalc.h > >> > >> [6] http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_varmacro.00/ > >> [7] http://alexkasko.bitbucket.org/jdk6/webrev_hotspot_varmacro.00.zip > >> [8] > >> http://hg.openjdk.java.net/jdk7u/jdk7u/hotspot/diff/bb74dc5ddf07/src/share/vm/trace/traceMacros.hpp > >> [9] http://alexkasko.bitbucket.org/jdk6/webrev_jdk_nio_typo.00/ > >> [10] http://alexkasko.bitbucket.org/jdk6/webrev_jdk_nio_typo.00.zip > >> [11] > >> http://hg.openjdk.java.net/jdk7u/jdk7u-dev/jdk/diff/8dd8266a2f4b/src/windows/native/sun/nio/ch/Net.c > >> [12] http://www.oracle.com/technetwork/community/oca-486395.html#k > >> > >> -- > >> Regards, > >> Alex Kasko > >> > > > > > -- > Regards, > Alex Kasko > Ok, I'll commit these for you. Do you have an OpenJDK username? Otherwise I'll just use the Contributed-by field. -- Andrew :) Free Java Software Engineer Red Hat, Inc. (http://www.redhat.com) PGP Key: 248BDC07 (https://keys.indymedia.org/) Fingerprint = EC5A 1F5E C0AD 1D15 8F1F 8F91 3B96 A578 248B DC07
