JDK-8037084 : -XX:+VerifyAliases is broken
  • Type: Bug
  • Component: hotspot
  • Sub-Component: compiler
  • Affected Version: 9,10,15
  • Priority: P4
  • Status: Open
  • Resolution: Unresolved
  • OS: generic
  • CPU: generic
  • Submitted: 2014-03-11
  • Updated: 2025-04-29
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.
Other
tbdUnresolved
Related Reports
Duplicate :  
Duplicate :  
Duplicate :  
Relates :  
Description
java -XX:+VerifyAliases -Xcomp -version results in a failing assert.
Comments
will get back to you on this soon.
18-08-2021

Hi [~jcm], do you have plans to fix this problem? Or is it worth fixing ? I've encountered the same problem and did a quick investigation, I think we can comment out this assertion: static void verify_memory_slice(const MergeMemNode* m, int alias_idx, Node* n) { if (!VerifyAliases) return; // don't bother to verify unless requested if (VMError::is_error_reported()) return; // muzzle asserts when debugging an error if (Node::in_dump()) return; // muzzle asserts when printing //assert(alias_idx >= Compile::AliasIdxRaw, "must not disturb base_memory or sentinel"); To allow verifying memory slice even it's a base memory
18-08-2021

Issue 1: assert(alias_idx >= Compile::AliasIdxRaw) failed: must not disturb base_memory or sentinel, soln: disable VerifyAliases for NoAlias case (i.e gen_stub code) Issue 2: assert(new_type->higher_equal_speculative(n->type())) failed: new type must refine old type solution will skip the assert for CountedLoop induction variable phinode. Issue 3: Error: assert((nat != __null) == (at != __null)) failed
14-06-2017

Hi Jamsheed, I'll assign this issue to you. If you don't want to investigate it, please assign it back to me. Thank you and best regards, Zoltan
23-11-2016

L(dev.flag is broken)L(rarely used flag, haven't been used since JDK6)H(none) = P4
11-04-2016

Restore.
10-09-2014

[~kvn], what do you think?
10-09-2014

This flag is broken since a very long time, i.e., it did not work in the first versions of 7. Remove or restore?
11-03-2014