Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
Alan, Michael, The subject line says it all! Nice to be able to remove this old code that is no longer necessary. Hopefully this will make maintenance a little easier going forward. http://cr.openjdk.java.net/~chegar/7030256/webrev.00/webrev/ Mike Duigou noticed I pushed this webrev to cr.openjdk yesterday and has already given feedback. Thanks Mike. -Chris.
Re: Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
Chris Hegarty wrote: Alan, Michael, The subject line says it all! Nice to be able to remove this old code that is no longer necessary. Hopefully this will make maintenance a little easier going forward. http://cr.openjdk.java.net/~chegar/7030256/webrev.00/webrev/ Mike Duigou noticed I pushed this webrev to cr.openjdk yesterday and has already given feedback. Thanks Mike. -Chris. This looks a good cleanup to me. Lots of 95/98 crud in the repository and always hard to get time to clear it out. I guess the only concern is that net.dll will now require iphlpapi.dll. Minor comments: src/windows/native/java/net/Inet6AddressImpl.c - you might need to re-aligned line 344. src/windows/native/java/net/net_util_md.c line - cast to jint??? -Alan.
Re: Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
On 03/24/11 10:55 AM, Alan Bateman wrote: Chris Hegarty wrote: Alan, Michael, The subject line says it all! Nice to be able to remove this old code that is no longer necessary. Hopefully this will make maintenance a little easier going forward. http://cr.openjdk.java.net/~chegar/7030256/webrev.00/webrev/ Mike Duigou noticed I pushed this webrev to cr.openjdk yesterday and has already given feedback. Thanks Mike. -Chris. This looks a good cleanup to me. Lots of 95/98 crud in the repository and always hard to get time to clear it out. I guess the only concern is that net.dll will now require iphlpapi.dll. Yes, we will now have a dependency on iphlpapi.dll, but this dll is part of Windows since Windows 2000 and the functions we are using are all available on XP. Does this address your concern, or do you have another issue? Minor comments: src/windows/native/java/net/Inet6AddressImpl.c - you might need to re-aligned line 344. Got it. Thanks. src/windows/native/java/net/net_util_md.c line - cast to jint??? I forgot to mention that I fixed a few compiler warning while I was here. Are you referring to the casting of SOCKET s to jint, line 982? SOCKET is an unsigned int pointer and needs an explicit cast. I thought it was safe, and is what is done in Java_sun_nio_ch_Net_socket0. -Chris.
Re: Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
Chris Hegarty wrote: Yes, we will now have a dependency on iphlpapi.dll, but this dll is part of Windows since Windows 2000 and the functions we are using are all available on XP. Does this address your concern, or do you have another issue? The concern is that we load yet another DLL at startup. It might be worth checking if it gets loaded already. : I forgot to mention that I fixed a few compiler warning while I was here. Are you referring to the casting of SOCKET s to jint, line 982? SOCKET is an unsigned int pointer and needs an explicit cast. I thought it was safe, and is what is done in Java_sun_nio_ch_Net_socket0. NET_Socket is declared to return an int so I assume you meant to cast to int rather than jint. -Alan.
Re: Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
On 03/24/11 11:49 AM, Alan Bateman wrote: Chris Hegarty wrote: Yes, we will now have a dependency on iphlpapi.dll, but this dll is part of Windows since Windows 2000 and the functions we are using are all available on XP. Does this address your concern, or do you have another issue? The concern is that we load yet another DLL at startup. It might be worth checking if it gets loaded already. You are right, before these changes iphlpapi.dll is not loaded unless NetworkInterface or ResolverConfigurationImpl are called. After these changes iphlpapi.dll is loaded once net.dll is loaded. I think we should use specify the loading of these dll's to be delayed. I can confirm that with the below changes iphlpapi.dll is only loaded when necessary. diff -r ef5bbbe0dd75 make/java/net/Makefile --- a/make/java/net/MakefileMon Mar 21 22:02:00 2011 -0700 +++ b/make/java/net/MakefileThu Mar 24 15:27:38 2011 + @@ -37,10 +37,6 @@ AUTO_FILES_JAVA_DIRS = java/net AUTO_FILES_JAVA_DIRS = java/net ifeq ($(PLATFORM), windows) -# Windows 9x module only needed on 32-bit build -ifeq ($(ARCH_DATA_MODEL), 32) - FILES_c += NetworkInterface_win9x.c -endif FILES_c += NTLMAuthSequence.c FILES_c += NetworkInterface_winXP.c else @@ -96,7 +92,9 @@ include $(BUILDDIR)/common/Library.gmk include $(BUILDDIR)/common/Library.gmk ifeq ($(PLATFORM), windows) - OTHER_LDLIBS = ws2_32.lib $(JVMLIB) + OTHER_LDLIBS = ws2_32.lib $(JVMLIB) \ + secur32.lib iphlpapi.lib delayimp.lib \ + /DELAYLOAD:secur32.dll /DELAYLOAD:iphlpapi.dll else OTHER_LDLIBS = $(LIBSOCKET) $(LIBNSL) -ldl $(JVMLIB) : I forgot to mention that I fixed a few compiler warning while I was here. Are you referring to the casting of SOCKET s to jint, line 982? SOCKET is an unsigned int pointer and needs an explicit cast. I thought it was safe, and is what is done in Java_sun_nio_ch_Net_socket0. NET_Socket is declared to return an int so I assume you meant to cast to int rather than jint. D'oh. Got it. Thanks, -Chris. -Alan.
Re: Code Review 7030256: Cleanup/Remove code supporting old Windows versions (98, NT, 2000, etc)
Chris Hegarty wrote: : You are right, before these changes iphlpapi.dll is not loaded unless NetworkInterface or ResolverConfigurationImpl are called. After these changes iphlpapi.dll is loaded once net.dll is loaded. I think we should use specify the loading of these dll's to be delayed. I can confirm that with the below changes iphlpapi.dll is only loaded when necessary. diff -r ef5bbbe0dd75 make/java/net/Makefile --- a/make/java/net/MakefileMon Mar 21 22:02:00 2011 -0700 +++ b/make/java/net/MakefileThu Mar 24 15:27:38 2011 + @@ -37,10 +37,6 @@ AUTO_FILES_JAVA_DIRS = java/net AUTO_FILES_JAVA_DIRS = java/net ifeq ($(PLATFORM), windows) -# Windows 9x module only needed on 32-bit build -ifeq ($(ARCH_DATA_MODEL), 32) -FILES_c += NetworkInterface_win9x.c -endif FILES_c += NTLMAuthSequence.c FILES_c += NetworkInterface_winXP.c else @@ -96,7 +92,9 @@ include $(BUILDDIR)/common/Library.gmk include $(BUILDDIR)/common/Library.gmk ifeq ($(PLATFORM), windows) - OTHER_LDLIBS = ws2_32.lib $(JVMLIB) + OTHER_LDLIBS = ws2_32.lib $(JVMLIB) \ + secur32.lib iphlpapi.lib delayimp.lib \ + /DELAYLOAD:secur32.dll /DELAYLOAD:iphlpapi.dll else OTHER_LDLIBS = $(LIBSOCKET) $(LIBNSL) -ldl $(JVMLIB) Thanks for doing this, this seems a good candidate for a delayed load. Thumbs up from me. -Alan
hg: jdk7/tl/jdk: 7029823: (ann) test/java/lang/annotation/package-info.java no longer compiles
Changeset: 632a96f5752d Author:jjh Date: 2011-03-24 11:40 -0700 URL: http://hg.openjdk.java.net/jdk7/tl/jdk/rev/632a96f5752d 7029823: (ann) test/java/lang/annotation/package-info.java no longer compiles Summary: Use @Deprecated instead of @java.lang.annotation.Documented Reviewed-by: jjg, smarks ! test/java/lang/annotation/PackageMain.java ! test/java/lang/annotation/package-info.java
hg: jdk7/tl/langtools: 6597678: JavaCompiler.getStandardFileManager always uses default charset not the one that user specifies
Changeset: 83260b3305ac Author:jjg Date: 2011-03-24 16:14 -0700 URL: http://hg.openjdk.java.net/jdk7/tl/langtools/rev/83260b3305ac 6597678: JavaCompiler.getStandardFileManager always uses default charset not the one that user specifies Reviewed-by: mcimadamore ! src/share/classes/com/sun/tools/javac/api/JavacTaskImpl.java ! src/share/classes/com/sun/tools/javac/api/JavacTool.java ! src/share/classes/com/sun/tools/javac/processing/JavacProcessingEnvironment.java ! src/share/classes/com/sun/tools/javac/util/JavacMessages.java + test/tools/javac/util/T6597678.java
hg: jdk7/tl/jdk: 7029680: fix test/sun/misc/Version/Version.java build parsing
Changeset: f326a018fd3f Author:smarks Date: 2011-03-24 17:26 -0700 URL: http://hg.openjdk.java.net/jdk7/tl/jdk/rev/f326a018fd3f 7029680: fix test/sun/misc/Version/Version.java build parsing Reviewed-by: ohair ! test/sun/misc/Version/Version.java
hg: jdk7/tl/jdk: 7023056: NPE from sun.security.util.ManifestEntryVerifier.verify during Maven build
Changeset: 4a64eefbfd7a Author:weijun Date: 2011-03-25 11:58 +0800 URL: http://hg.openjdk.java.net/jdk7/tl/jdk/rev/4a64eefbfd7a 7023056: NPE from sun.security.util.ManifestEntryVerifier.verify during Maven build Reviewed-by: mullan ! src/share/classes/java/util/jar/JarVerifier.java ! src/share/classes/sun/security/util/ManifestEntryVerifier.java + test/java/util/jar/JarFile/MevNPE.java
Re: HttpCookie.domainMatches("hostname.local", "hostname") return false
Hi Chris, Is there any progress on this issue? -- Best Regards, Sean Chou
