RFR: 8169317: [s390] Various minor bug fixes and adaptions.

[Lindenmaier, Goetz]

Hi,

I updated the webrev with a fix by Martin Doerr, that fixes
an assertion showing on single processor machines.
Could I please get reviews?
http://cr.openjdk.java.net/~goetz/wr16/8169317-s390_fixes/webrev-02/

@Shahid, this contains the fix Martin sent you, could you confirm that this
helps with your issue?

Best regards,
  Goetz.

> -----Original Message-----
> From: Lindenmaier, Goetz
> Sent: Montag, 7. November 2016 12:53
> To: hotspot-compiler-dev at openjdk.java.net
> Cc: Cindy Lee <cinderel at ca.ibm.com>; Amod Borkar
> <aborkar at us.ibm.com>; Shahid Shaikh <shahid at us.ibm.com>
> Subject: RFR: 8169317: [s390] Various minor bug fixes and adaptions.
> 
> Hi,
> 
> 
> 
> This change contains a row of minor fixes to the s390 port.
> 
> Please review this change:
> 
> http://cr.openjdk.java.net/~goetz/wr16/8169317-s390_fixes/webrev-01/
> 
> 
> 
> The fixes in detail:
> 
> 
> s390.ad
> Fix rule indexOfChar_U.
> 
> c1_LIRAssembler_s390.cpp
> Register overwritten in heap based compressed oops mode. We never see
> this mode on our systems, but it is a possible configuration.
> 
> templateTable_s390.cpp, nativeInst_s390.cpp:
> Wrong register passed as argument to call_VM
> 
> frame_s390.cpp, frame_s390.inline.hpp. os_linux_s390.cpp
> Some adaptions for "8151956: Support non-continuous CodeBlobs in
> HotSpot"
> 
> Remove flag UseCountLeadingZerosInstruction. This instruction is available
> on all supported s390 cpus, and the tests for this flag depend on a string in
> the x86 cpu feature list.
> 
> vm_version_s390.cpp
> Add 'aes' and 'sha' to cpu feature string if the corresponding extensions are
> implemented on the current cpu. Needed for jtreg tests.
> 
> os_linux_s390.cpp
> Fix frame popping for debug / fastdebug build. Else NMT tests fail.
> 
> 
> 
> Best regards,
> 
>   Goetz.