JDK-8207252 : C1 still does eden allocations when TLAB is enabled
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 11
  • Priority: P3
  • Status: Resolved
  • Resolution: Fixed
  • Submitted: 2018-07-13
  • Updated: 2018-08-09
  • Resolved: 2018-07-23
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 12
11 b24Fixed 12Fixed
Related Reports
Relates :  
JDK-8190862 - Interpreter never refills TLAB
Relates :  
JDK-8208778 - tests crash in _platform_memmove$VARIANT$Nehalem+0x60 on MacOSX
Relates :  
JDK-8206936 - Tests serviceability/jvmti/HeapMonitor/MyPackage/HeapMonitor* doesn't work with additional VM options
Description
JDK-8190862 fixed the interpreter to not do eden allocations when TLAB is enabled.

C1 still does it and this bug addresses that
Comments
RFR thread links - http://mail.openjdk.java.net/pipermail/hotspot-runtime-dev/2018-July/029195.html http://mail.openjdk.java.net/pipermail/hotspot-runtime-dev/2018-July/029236.html
19-07-2018
initial ILW = similar to related JDK-8190862! = P3
18-07-2018
Okay, changed the priority and targeted to 11.
16-07-2018
I think it is a low risk. It changes a flag that handles eden allocations if possible which is off by default for G1, the default GC for 11 anyway. So, from my understanding, this is a NOP in the general case, only corner cases would perhaps see a difference.
16-07-2018
Is it risky to fix in 11? If the risk is small then it is better fix in 11 - I can adjust the priority.
16-07-2018
I let you all decide the priority :) If this does not go in 11 for the RDP1 phase, then for JDK-8206936, I have to disable testing with TieredStopAtLevel. We could do that, do this fix for 12 and then relax the testing requirements. Whatever you all prefer :)
14-07-2018
What is an intention on the FixVersion for this bug? If it is for 11 then P4 is too low for the RDP1 phase of the release.
14-07-2018