RFR: 8267926: AsyncLogGtest.java fails on assert with: decorator was not part of the decorator set specified at creation. [v7]

Xin Liu xliu at openjdk.java.net
Thu Jun 3 07:58:37 UTC 2021


On Thu, 3 Jun 2021 07:35:11 GMT, Xin Liu <xliu at openjdk.org> wrote:

>> The root cause of the intermittent failure is that _decorators in LogDecorations 
>> may be inconsistent with LogOutput::_decorators. It could happen when gtest disables a
>> Log output via set_log_config(TestLogFileName, "all=off").
>> 
>> Since we copy the entire logDecorations, it's reasonable to copy _decorator as well. 
>> LogDecorator is a bitmask of uint. It's even smaller than a reference on LP64 platforms.
>> 
>> This patch not only fixes intermittent crash in JDK-8267926, but also fixes JDK-8267952.
>> Two hazards of async logging in `LogConfiguration::configure_output` are eliminated. 
>> This patch also adds a storestore barrier to make sure it's safe on weak consistency machines.
>
> Xin Liu has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Use RAII to protect _rotation_semaphore.

I double down that one flush() can eliminate two hazards. 2 concurrent tests are provided to demonstrate `LogConfiguration::configure_output` is MT-safe. 

Previously,  the test was in bad shape which cause unnecessary rotation and quite big(100M+) on /tmp. I guess that's the root cause of failure on Linux/aarch64. patched.

This patch not only fixes intermittent crash in JDK-8267926, but also fixes JDK-8267952.

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

PR: https://git.openjdk.java.net/jdk/pull/4257


More information about the hotspot-runtime-dev mailing list