RFR: 8317371: GenShen: Needs improved support for humongous allocation [v3]

[William Kemper]

On Wed, 11 Oct 2023 19:09:52 GMT, Kelvin Nilsen <kdnilsen at openjdk.org> wrote:

>> Several improvements:
>> 1. If there is a humongous allocation failure, go immediately to Full GC rather than trying 64 degenerated GC first
>> 2. Reduce likelihood of humongous allocation errors by:
>>     a. Allocating regular objects from top end of memory while allocating humongous from bottom end of memory
>>     b. (Especially) in the case that regular regions that are promoted in place reside within the bottom end of memory, detect that old-gen memory has become fragmented and trigger old-gen GC, which will make a special effort to move these out-of-place old-gen regions to higher addresses.
>
> Kelvin Nilsen has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Cosmetic improvements

src/hotspot/share/gc/shenandoah/shenandoahFreeSet.cpp line 570:

> 568:     case ShenandoahAllocRequest::_alloc_shared: {
> 569:       // Try to allocate in the mutator view
> 570:       // Allocate within mutator free from high memory to low so as to preserve low memory for humongous allocations

> a. Allocating regular objects from top end of memory while allocating humongous from bottom end of memory

Don't we usually take the collector reserve from the top end of the heap? Would it make sense to assume humongous objects tend to be long lived and allocate them from the back of the heap?

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

PR Review Comment: https://git.openjdk.org/shenandoah/pull/336#discussion_r1357209856