JDK-8230671 : x86_32 build failures after JDK-8229496
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 11,14
  • Priority: P1
  • Status: Resolved
  • Resolution: Fixed
  • Submitted: 2019-09-05
  • Updated: 2022-02-06
  • Resolved: 2019-09-06
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 11 JDK 13 JDK 14
11.0.6-oracleFixed 13.0.3Fixed 14 b14Fixed
Related Reports
Relates :  
Description
Apparently, rRegL is incorrect operand name for x86_32.ad.

* For target hotspot_variant-server_support_adlc__adlc_run_exec:
/home/shade/trunks/jdk-jdk/src/hotspot/cpu/x86/x86_32.ad(7318) Syntax Error: :undefined operand type rRegL

Error Context:  >>> <<<dst ) %{
/home/shade/trunks/jdk-jdk/src/hotspot/cpu/x86/x86_32.ad(7319) Syntax Error: :missing '%{' in instruction definition

Error Context:  >>>m<<<atch(Set dst (CastLL dst));
/home/shade/trunks/jdk-jdk/src/hotspot/cpu/x86/x86_32.ad(7320) Syntax Error: :expected one of - instruct, operand, ins_attrib, op_attrib, source, register, pipeline, encode
     Found format
Error Context:  >>> <<<%{ "#castLL of $dst" %}
/home/shade/trunks/jdk-jdk/src/hotspot/cpu/x86/x86_32.ad(7321) Syntax Error: :expected one of - instruct, operand, ins_attrib, op_attrib, source, register, pipeline, encode
     Found ins_encode
Error Context:  >>>(<<< /*empty encoding*/ );
/home/shade/trunks/jdk-jdk/src/hotspot/cpu/x86/x86_32.ad(7322) Syntax Error: :expected one of - instruct, operand, ins_attrib, op_attrib, source, register, pipeline, encode
     Found ins_cost
Error Context:  >>>(<<<0);
   ... (rest of output omitted)
Comments
Note, however, that TestDivZeroCheckControl.java from the original fix does consistently timeout on 32-bit JDK. It's hard to find a real 32-bit Windows system but a run with 32bit JDK on a 64bit system always does only some 950 iterations out of 50_000 in 2 minutes (all the same on Hyper-V, VirtualBox, and a real machine). With 64bit JDK the whole test run takes just few seconds. Now, could this degrade performance on 32bit Windows? 32bit Linux systems seem OK. (Update: answer seems No: ) Ekaterina Vergizova tried it with and without the fixes and reported consistent slow behavior.
24-03-2020

Fix Request for 13u The patch fixes a build failure caused by JDK-8229496. Applies cleanly to 13u.
24-03-2020

11u Fix Request This fixes a build failure from JDK-8229496 for which I requested the backport. Patch applies cleanly to 11u. tier1 tests pass with the patch.
02-10-2019

URL: https://hg.openjdk.java.net/jdk/jdk/rev/66c880e5d21d User: shade Date: 2019-09-06 07:26:37 +0000
06-09-2019

ILW = HMH = P1
06-09-2019

RFR: https://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2019-September/034954.html
05-09-2019