RFR: 8341334: CDS: Parallel relocation [v11]

[Claes Redestad]

On Wed, 6 Nov 2024 18:16:44 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

>> In Leyden performance investigations, we have figured that `ArchiveRelocationMode=0` is saving 5..7 ms on HelloWorld startup. Mainline defaults to `ARM=1`, _losing_ as much. `ARM=0` was switched to `ARM=1` with [JDK-8294323](https://github.com/openjdk/jdk/commit/be70bc1c58eaec876aa1ab36eacba90b901ac9b8), which was delivered to JDK 17+ in in Apr 2023.
>> 
>> Profiling shows we spend time mem-faulting the memory loading the RO/RW regions, about 15 MB total. 15 MB in 5ms amounts to >4GB/sec, close to the single-threaded limits. I suspect the impact is larger if we relocate larger Metaspace, e.g. after dumping a CDS archive from a large application. There is little we can do to make the actual relocation part faster: the overwhelming majority of samples is on kernel side.
>> 
>> This PR implements the infrastructure for fast archive workers, and leverages it to perform parallel core regions relocation. The RW/RO regions this code traverses is large, and we can squeeze more performance by parallelizing it. 
>> 
>> (I'll put some performance data in the comments)
>> 
>> Additional testing:
>>  - [x] Linux x86_64 server fastdebug, `runtime/cds`
>>  - [ ] Linux AArch64 server fastdebug, `all`
>
> Aleksey Shipilev has updated the pull request incrementally with two additional commits since the last revision:
> 
>  - Make sure archive workers are shutdown
>  - Cosmetic changes

src/hotspot/share/cds/archiveUtils.cpp line 444:

> 442: }
> 443: 
> 444: void ArchiveWorkers::start_worker_if_needed() {

Not sure it matters, but when `_started_workers` is or becomes equal to `_num_workers` we could pass that bit of information on to avoid calling `start_worker_if_needed`. Might avoid a possible burst of contended reads.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/21302#discussion_r1831856479