Re: [PATCH v3 02/12] sparse-index: include EXTENDED flag when expanding

> On Tue, May 18, 2021 at 7:57 AM Derrick Stolee <stolee@gmail.com> wrote:
>>
>> On 5/17/2021 9:33 PM, Elijah Newren wrote:
>>> Is it possible that the setting of CE_EXTENDED is just a workaround
>>> that happens to force the index to be written in cases where the logic
>>> is otherwise thinking it can get away without one?  Or is there
>>> something I'm missing about why the CE_EXTENDED flag is actually
>>> needed here?
>>
>> This is happening within the context of ensure_full_index(), so we
>> are creating new cache entries and want to mimic what they would
>> look like on-disk. Something within do_write_index() discovers that
>> since CE_SKIP_WORKTREE is set, then also CE_EXTENDED should be set
>> in order to ensure that the on-disk representation has enough room
>> for the CE_SKIP_WORKTREE bit.
> 
> Yeah, I think it's this part:
> 
>         /* reduce extended entries if possible */
>         cache[i]->ce_flags &= ~CE_EXTENDED;
>         if (cache[i]->ce_flags & CE_EXTENDED_FLAGS) {
>             extended++;
>             cache[i]->ce_flags |= CE_EXTENDED;
>         }
> 
>>
>> I suppose this might not have a meaningful purpose other than when
>> I compare a full index against an expanded sparse-index and check
>> if their flags match.
> 
> Ah, you're just setting this flag in advance of do_write_index() being
> called so that you can compare in memory values and check they match
> without doing a write-to-disk-and-read-back cycle.  Makes sense, but
> it'd be nice to see this in the commit message.