RFR: 8322296: Introduce CardWord for word iteration on CardTable [v2]

[Albert Mingkun Yang]

On Thu, 21 Dec 2023 21:13:45 GMT, Kim Barrett <kbarrett at openjdk.org> wrote:

>> Albert Mingkun Yang has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains three additional commits since the last revision:
>> 
>>  - review
>>  - Merge branch 'master' into card-word
>>  - card-word
>
> src/hotspot/share/gc/g1/g1RemSet.cpp line 559:
> 
>> 557: 
>> 558:     CardValue* find_first_dirty_card(CardValue* i_card) const {
>> 559:       while (!CardTable::is_word_aligned(i_card)) {
> 
> pre-existing: This loop assumes initial i_card < _end_card and _end_card is word aligned.  Should be
> asserted. Similarly in find_first_non_dirty_card.

It's asserted in the constructor.

> src/hotspot/share/gc/serial/cardTableRS.cpp line 365:
> 
>> 363:   // Word comparison
>> 364:   while (current_card + sizeof(CardWord) <= end_card) {
>> 365:     CardWord* current_word = reinterpret_cast<CardWord*>(current_card);
> 
> When the idea of something like CardWord was suggested, a named accessor to package this was part
> of that suggestion.

I find that too "caller-specific" to be included inside `CardTable`. Creating an abstraction here breaks the flow, so I kept it as is.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/17144#discussion_r1434532937
PR Review Comment: https://git.openjdk.org/jdk/pull/17144#discussion_r1434534227