RFR: 8333658: NMT: Use an allocator with 4-byte pointers to save memory in NativeCallStackStorage [v32]

[Thomas Stuefe]

On Mon, 24 Jun 2024 14:12:49 GMT, Johan Sjölen <jsjolen at openjdk.org> wrote:

>> Hi,
>> 
>> This PR introduces a new allocator, `IndexedFreelistAllocator`. It uses a `GrowableArray` in order to have 4-byte "pointers" to its elements and also works as a freelist as unused slots form a linked list. This allocator cannot shrink its used memory. I'm always open for better names.
>> 
>> We then use this allocator in order to store the `NativeCallStackStorage` hash table. This saves `4 * 4099 = ~16KiB` of memory as each table element is now only 4 bytes. For each stored stack it also saves 4 bytes. The fact that the allocator cannot shrink is fine, we do not ever remove stacks from the storage.
>> 
>> The main point of this is to introduce a pointer-generic experimential API, so I also implemented CHeap and Arena allocators. It's currently placed in NMT, but we might want to move it into utilities. It uses a bit of template-magic, but my IDE (Emacs+clangd) was actually able to figure things out when the types didn't line up correctly etc, so it's not an enemy to IDE help.
>> 
>> It sounded expensive to have the GrowableArray continously realloc its underlying data array, so I did a basic test where we allocate 1 000 000 stacks and push them into NativeCallStackStorage backed by different allocators. This is available in the PR.
>> 
>> The results are as follows on linux-x64-slowdebug:
>> 
>> 
>> Generate stacks... Done
>> Time taken with GrowableArray: 8341.240945
>> Time taken with CHeap: 12189.031318
>> Time taken with Arena: 8800.703092
>> Time taken with GrowableArray again: 8295.508829
>> 
>> 
>> And on linux-x64:
>> 
>> 
>> Time taken with GrowableArray: 8378.018135
>> Time taken with CHeap: 12437.347868
>> Time taken with Arena: 8758.064717
>> Time taken with GrowableArray again: 8391.076291
>> 
>> 
>> Obviously, this is a very basic benchmark, but it seems like we're faster than CHeap and Arena for this case.
>
> Johan Sjölen has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Rename tests

src/hotspot/share/nmt/nmtNativeCallStackStorage.hpp line 76:

> 74:   // 4099 gives a 50% probability of collisions at 76 stacks (as per birthday problem).
> 75:   static const constexpr int default_table_size = 4099;
> 76:   int _table_size;

Pre-existing. Both table size and, arguably, the table itself are never changed, since we don't resize. At least size could be const, _table too if you move allocation to init list. (latter is a matter of taste, up to you).

src/hotspot/share/nmt/nmtNativeCallStackStorage.hpp line 108:

> 106:       }
> 107:     }
> 108:   }

Could you move this, and the dtor, over to the cpp file, too? You probably can remove the allocation.hpp include then. Less include deps are always good.

test/hotspot/gtest/nmt/test_arrayWithFreeList.cpp line 71:

> 69:     return e;
> 70:   }
> 71: };

nit, add newline

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

PR Review Comment: https://git.openjdk.org/jdk/pull/18979#discussion_r1652713763
PR Review Comment: https://git.openjdk.org/jdk/pull/18979#discussion_r1652764983
PR Review Comment: https://git.openjdk.org/jdk/pull/18979#discussion_r1652769018