Bad push for JDK-8230728: Thin stroked shapes are not rendered if affine transform has flip bit

[Langer, Christoph]

Hi Goetz, Laurent,

I had a look to the AIX test failures that we observe. It seems more likely to me that it is not related to the push of 8230728 but to another issue with fontconfig on AIX which we're chasing for several weeks now. So technically I think there's nothing wrong with the push.

But since we're in the CPU phase of 11.0.5 the change should be reverted in 11u and be pushed to 11u-dev.

So, @Laurent, please do the backout as outlined by Goetz and push your change to jdk11u-dev as well.

Thanks
Christoph

> -----Original Message-----
> From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net> On
> Behalf Of Lindenmaier, Goetz
> Sent: Dienstag, 1. Oktober 2019 08:44
> To: Laurent Bourgès <bourges.laurent at gmail.com>; jdk-updates-
> dev at openjdk.java.net
> Subject: [CAUTION] RE: Bad push for JDK-8230728: Thin stroked shapes are
> not rendered if affine transform has flip bit
> 
> Hi Laurent,
> 
> please open a JBS Issue
>   'Backout "8230728: Thin stroked shapes are not rendered if affine transform
> has flip bit" from jdk11u'
> and send a RFR with the backout change. Tag it as jdk11u-critical-request.
> 
> Also, I see crashes in our tests that were triggered by your
> push. I didn't investigate, but they are in the fonts area.
> 
> The following jtreg tests crashed on AIX:
> javax/swing/Headless/HeadlessJMenuBar.java
> javax/swing/Headless/HeadlessJSlider.java
> javax/swing/Headless/HeadlessJTextArea.java
> javax/swing/Headless/HeadlessJToolBar_Separator.java
> javax/swing/Headless/HeadlessJToolBar.java
> 
> #
> # A fatal error has been detected by the Java Runtime Environment:
> #
> #  SIGBUS (0xa) at pc=0x0900000006f917b8, pid=13959206, tid=5141
> #
> # JRE version: OpenJDK Runtime Environment (11.0.5) (build 11.0.5-
> internal+0-adhoc.openjdk.jdk11u)
> # Java VM: OpenJDK 64-Bit Server VM (11.0.5-internal+0-
> adhoc.openjdk.jdk11u, mixed mode, tiered, compressed oops, g1 gc, aix-
> ppc64)
> # Problematic frame:
> # C  0x0900000006f917b8
> #
> 
> 0x0000000116b081d0 - 0x0900000006f8a094 (unknown module)::(unknown
> function)+?
> 0x0000000116b08260 - 0x0900000006f8a150 (unknown module)::(unknown
> function)+?
> 0x0000000116b082d0 - 0x0900000006f8ace0 (unknown module)::(unknown
> function)+?
> 0x0000000116b08340 - 0x090000000482e79c
> libawt_headless.so::Java_sun_font_FontConfigManager_getFontConfig+0x6
> 1c  (C saves_cr saves_lr stores_bc gpr_saved:18 fixedparms:6 )
> 0x0000000116b08580 - 0x0a0001000000ed7c (unknown module)::(unknown
> function)+?
> 0x0000000116b086a0 - 0x0a000100000085a8 (unknown module)::(unknown
> function)+?
> 0x0000000116b08770 - 0x0a000100000085a8 (unknown module)::(unknown
> function)+?
> 0x0000000116b08870 - 0x0a00010000008300 (unknown module)::(unknown
> function)+?
> 0x0000000116b08930 - 0x0a000100000009cc (unknown module)::(unknown
> function)+?
> 0x0000000116b08ae0 - 0x09000000050f9598
> libjvm.so::JavaCalls::call_helper(JavaValue*,const
> methodHandle&,JavaCallArguments*,Thread*)+0x438  (C++ saves_lr
> stores_bc gpr_saved:15 fixedparms:4 )
> 0x0000000116b08ca0 - 0x09000000049bb2f8
> libjvm.so::os::os_exception_wrapper(void(*)(JavaValue*,const
> methodHandle&,JavaCallArguments*,Thread*),JavaValue*,const
> methodHandle&,JavaCallArguments*,Thread*)+0x38  (C++ saves_lr
> stores_bc fixedparms:5 )
> 0x0000000116b08d10 - 0x09000000050fc834
> libjvm.so::JavaCalls::call(JavaValue*,const
> methodHandle&,JavaCallArguments*,Thread*)+0x34  (C++ saves_lr
> stores_bc fixedparms:4 )
> 0x0000000116b08d80 - 0x090000000494e994
> libjvm.so::JVM_DoPrivileged+0xc14  (C++ saves_lr stores_bc gpr_saved:14
> fixedparms:5 )
> 0x0000000116b09750 - 0x090000000479a318
> libjava.so::Java_java_security_AccessController_doPrivileged__Ljava_securit
> y_PrivilegedAction_2+0x18  (C saves_lr stores_bc fixedparms:3 )
> 0x0000000116b097c0 - 0x0a0001000000ed7c (unknown module)::(unknown
> function)+?
> 0x0000000116b098e0 - 0x0a00010000008300 (unknown module)::(unknown
> function)+?
> 
> Crash on AIX in an SAP application we test:
> 
> 0x0000000119327d30 - 0x0900000006f8a150 (unknown module)::(unknown
> function)+?
> 0x0000000119327da0 - 0x0900000006fb1970 (unknown module)::(unknown
> function)+?
> 0x0000000119327e20 - 0x0900000006f914e8 (unknown module)::(unknown
> function)+?
> 0x0000000119327eb0 - 0x0900000006914ee4
> libawt_headless.so::getFontConfigLocations+0x1c4  (C saves_lr stores_bc
> gpr_saved:18 )
> 0x0000000119327fc0 - 0x09000000069146c0
> libawt_headless.so::getPlatformFontPathChars+0x20  (C saves_lr stores_bc
> gpr_saved:3 fixedparms:3 )
> 0x0000000119328050 - 0x0900000006914618
> libawt_headless.so::Java_sun_awt_FcFontManager_getFontPathNative+0x7
> 8  (C saves_lr stores_bc gpr_saved:1 fixedparms:4 )
> 0x00000001193280e0 - 0x0a00010028fb28ec (unknown module)::(unknown
> function)+?
> 0x00000001193281a0 - 0x0a00010003c21b88 (unknown module)::(unknown
> function)+?
> 
> So maybe we should investigate this before pushing to jdk11u-dev.
> 
> Best regards,
>   Goetz.
> 
> 
> 
> > -----Original Message-----
> > From: jdk-updates-dev <jdk-updates-dev-bounces at openjdk.java.net> On
> > Behalf Of Laurent Bourgès
> > Sent: Montag, 30. September 2019 18:13
> > To: jdk-updates-dev at openjdk.java.net
> > Subject: Bad push for JDK-8230728: Thin stroked shapes are not rendered if
> > affine transform has flip bit
> >
> > Hi,
> >
> > I pushed by mistake on the jdk11u (11.0.5) repository, not on jdk11u-dev !
> > Sorry, sorry, sorry.
> > Previously I only had to deal with 10u, 12u...
> >
> > See https://bugs.openjdk.java.net/browse/JDK-8231621
> >
> > How can I (or you) revert that changeset ? What is the process ?
> >
> > PS: I will get rid of my local copy jdk11u (master), I already retrieved
> > the jdk11u-dev to use instead.
> >
> > Thanks for your help,
> > Laurent