RFR [XS]: 8234070: solaris sparc build fails in templateInterpreterGenerator.cpp - was RE: fastdebug Solaris build error in jdk/jdk

Harold Seigel harold.seigel at oracle.com
Wed Nov 13 14:08:29 UTC 2019


Hi Matthias,

I was able to successfully build Solaris Sparc with the latest sources.  
Can you see if you still have this problem?

It looks like Solaris Sparc could find the definition of verify_FPU() in 
hotspot/src/cpu/sparc/interp_masm_sparc.hpp:  void verify_FPU(int 
stack_depth, TosState state = ftos) {}      // No-op.

Thanks, Harold

On 11/13/2019 5:02 AM, Baesken, Matthias wrote:
> Hello,  please review the following change .
>
> It fixes the fastdebug  Solaris Sparc build after recent changes     8233918   / 8233498 .
>
> Erik, may I add you as a reviewer ?
>
>
>
> Bug/webrev:
> https://bugs.openjdk.java.net/browse/JDK-8234070
>
> http://cr.openjdk.java.net/~mbaesken/webrevs/8234070.0/
>
> Thanks, Matthias
>
>
> From: Erik Österlund <erik.osterlund at oracle.com>
> Sent: Mittwoch, 13. November 2019 09:41
> To: Baesken, Matthias <matthias.baesken at sap.com>; patric.hedlin at oracle.com
> Cc: Zeller, Arno <arno.zeller at sap.com>; Langer, Christoph <christoph.langer at sap.com>
> Subject: Re: fastdebug Solaris build error in jdk/jdk
>
> Hi Matthias,
>
> Your suggestion sounds reasonable.
>
> Thanks,
> /Erik
> On 2019-11-13 09:29, Baesken, Matthias wrote:
> Hello,  I just opened
>
> https://bugs.openjdk.java.net/browse/JDK-8234070
>
> 8234070: solaris sparc build fails in templateInterpreterGenerator.cpp
>
> After 8233498: Remove dead code. ,  and    8233918: 8233498 broke build on SPARC
>    the (fast)debug  Solaris build is still broken , error is
>
> /nightly/jdk/src/hotspot/share/interpreter/templateInterpreterGenerator.cpp", line 376: Error: verify_FPU is not a member of InterpreterMacroAssembler.
> 1 Error(s) detected.
>
>
> My suggestion would be  to remove   the  verify_FPU  call  from  templateInterpreterGenerator.cpp   because it is a no-op  anyway these  days on almost all platforms and  of  very limited use .
>
>
>
> Best regards, Matthias
>
>
>
>


More information about the hotspot-dev mailing list