On Tue, 28 Apr 2026 06:56:15 GMT, Marc Chevalier <[email protected]> wrote:

>> `acmp` for value object might be expensive: it may need substitutability 
>> test, which is done through a Java call. That is the last resort when 
>> everything failed, that is: we are in the presence of two value objects of 
>> the same class, not statically known or speculated (profile absent or 
>> polluted). While a lot of the acmp logic is trying to improve acmp at 
>> compile-time, this change proposes a runtime check to take a fast path.
>> 
>> The idea is that if the object is small enough, we can read all the fields 
>> at once with a single 8-byte load, we exclude the non-interesting bits with 
>> a mask, and we can simply compare the masked value from both objects.
>> 
>> The first step happens at class loading. When the class is loaded, we check 
>> whether all the fields can be read at once, from which offset, and the mask 
>> to filter out useless bits. Useless bits may have two sources:
>> - padding between fields
>> - header: this happens because we cannot necessarily start reading at 
>> `payload_offset`, the payload might be smaller than a long, and we could 
>> overread. On the other hand, the header is always long enough (never less 
>> than 8 bytes), so we may read from somewhere in the header to make sure we 
>> don't go further than the payload. This piece of header must be filtered out.
>> 
>> This strategy may not apply if the object is too big, or if it contains an 
>> oop. Loading oop without precaution, concurrently with the GC is not a good 
>> idea. This limitation is rather harmless: this fast path is mostly made not 
>> to pay the price of a complicated `acmp` for magically migrated classes 
>> (such as `Integer`), which are small and don't contain oops.
>> 
>> There is an interesting corner case: the mask can be 0 for an empty value 
>> class. We use a negative offset to signal that the fast path doesn't apply, 
>> since it is not a valid value for the offset: we shouldn't load before the 
>> object.
>> 
>> The second step is to prepare the runtime check. In `do_acmp`, when we used 
>> to prepare the Java call, now we have a parallel fast path. The condition 
>> uses the fact that we already have the class node available (used for 
>> checking operands have the same types). We can get the fast acmp offset from 
>> the class, check if it looks valid (that is non-negative), and if so, take 
>> the fast path that will load the mask, do the two loads and masking, and 
>> compare the given values. If the class of the operand is known precisely at 
>> parsing, we don't emit the fast path since the call will be nicely 
>> intrinsified.
>> 
>> The thir...
>
> Marc Chevalier has updated the pull request with a new target base due to a 
> merge or a rebase. The pull request now contains 31 commits:
> 
>  - Move the flag around
>  - Merge remote-tracking branch lworld into JDK-8381472
>  - Printing is gone again
>  - Remove the old way
>  - Fix asserts
>  - Let new version do the job
>  - Flag + using _layout_info
>  - Printing is back
>  - cosmetics
>  - Typo in acmp
>  - ... and 21 more: 
> https://git.openjdk.org/valhalla/compare/cfda30f3...88740943

src/hotspot/share/classfile/classFileParser.cpp line 5376:

> 5374: 
> 5375:   for (int i = 0; i < _layout_info->_nonoop_acmp_map->length(); i++) {
> 5376:     int field_start = _layout_info->_nonoop_acmp_map->at(i)._offset - 
> _layout_info->_payload_offset;

The entries in the acmp_maps are not fields, they are just pieces of contiguous 
memory used by one or several fields. An entry can cover multiple fields, or a 
field can be split over several entries. Renaming the local variables in this 
method avoiding the term "field" would prevent confusion.

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

PR Review Comment: 
https://git.openjdk.org/valhalla/pull/2353#discussion_r3156812800

Reply via email to