RFR: 8286876: NMT.test_unaliged_block_address_vm_assert fails if using clang toolchain [v3]

Johan Sjölen jsjolen at openjdk.org
Fri Dec 9 11:39:02 UTC 2022


On Thu, 8 Dec 2022 19:18:07 GMT, Thomas Stuefe <stuefe at openjdk.org> wrote:

>> Johan Sjölen has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains four commits:
>> 
>>  - Comments
>>  - Merge remote-tracking branch 'origin/master' into JDK-8286876
>>  - Generate assert_block_integrity thru templated function
>>  - Fix the bug
>
> src/hotspot/share/services/mallocHeader.cpp line 36:
> 
>> 34: 
>> 35: 
>> 36: void MallocHeader::print_block_on_error(outputStream* st, address bad_address) {
> 
> Just noticed, this seems wrong. As I wrote before, print_block_on_error() should take two pointers, the header and the corruption point. The header before was implied by `this`, now that you made it a static method you need to pass the method pointer explicitly. 
> 
> That said, what was the reason again this was made static? We only call this now for cases where the header pointer is valid.

Good job noticing this, I've been staring at this PR for too long :-).

This was a refactoring made because of an old regression, which you fixed in PR #11472. I'll take in your fix by merging with master.

-------------

PR: https://git.openjdk.org/jdk/pull/11465


More information about the hotspot-runtime-dev mailing list