RFR(L): 8145117: PPC64: Remove cpp interpreter implementation

Doerr, Martin martin.doerr at sap.com
Tue Dec 15 11:48:31 UTC 2015


Hi Götz,

thank you for doing this change. It looks good.
The only thing I'm missing is the description about the frame layout which was in the CC_INTERP part, but also describes frames which we still have. It would be nice if you could change the comments in frame_ppc.hpp like this:

  //  Frame layout for the Java template interpreter on PPC64.
  //
  //  In these figures the stack grows upwards, while memory grows
  //  downwards. Square brackets denote regions possibly larger than
  //  single 64 bit slots.
  //
  //  STACK (interpreter is active):
  //    0       [TOP_IJAVA_FRAME]
  //            [PARENT_IJAVA_FRAME]
  //            ...
  //            [PARENT_IJAVA_FRAME]
  //            [ENTRY_FRAME]
  //            [C_FRAME]
  //            ...
  //            [C_FRAME]
  //
  //  With the following frame layouts:
  //  TOP_IJAVA_FRAME:
  //    0       [TOP_IJAVA_FRAME_ABI]
  //            alignment (optional)
  //            [operand stack]
  //            [monitors] (optional)
  //            [IJAVA_STATE]
  //            note: own locals are located in the caller frame.
  //
  //  PARENT_IJAVA_FRAME:
  //    0       [PARENT_IJAVA_FRAME_ABI]
  //            alignment (optional)
  //            [callee's Java result]
  //            [callee's locals w/o arguments]
  //            [outgoing arguments]
  //            [used part of operand stack w/o arguments]
  //            [monitors]      (optional)
  //            [IJAVA_STATE]
  //
  //  ENTRY_FRAME:
  //    0       [PARENT_IJAVA_FRAME_ABI]
  //            alignment (optional)
  //            [callee's Java result]
  //            [callee's locals w/o arguments]
  //            [outgoing arguments]
  //            [ENTRY_FRAME_LOCALS]


Thanks and best regards,
  Martin

-----Original Message-----
From: hotspot-runtime-dev [mailto:hotspot-runtime-dev-bounces at openjdk.java.net] On Behalf Of Coleen Phillimore
Sent: Donnerstag, 10. Dezember 2015 21:51
To: hotspot-runtime-dev at openjdk.java.net
Subject: Re: RFR(L): 8145117: PPC64: Remove cpp interpreter implementation


Hi Goetz,

This looks great!  I have all but the formatting changes in the change 
that I'm working on also, and have the equivalent changes to the other 
platforms.

There are shared code changes that #include cppInterpreter_ppc.cpp that 
may prevent Zero from being built on ppc.  I will remove these in my 
upcoming patch.

The ppc implementation of the cppInterpreter was the only one that 
seemed to be close to complete (or was the most recently working 
implementation).  The others will are more clear that they are bit 
rotten and should be removed.

Thanks!
Coleen


On 12/10/15 10:35 AM, Lindenmaier, Goetz wrote:
> Hi,
>
> This change removed the cpp interpreter form the ppc sources.
> It's broken anyways.  The change is ppc only.
> http://cr.openjdk.java.net/~goetz/webrevs/8145117-ccInterp/webrev.00/
>
> Best regards,
>    Goetz.
>
>



More information about the hotspot-runtime-dev mailing list