On 3/17/2022 11:55 AM, Victoria Dye via GitGitGadget wrote:
> From: Victoria Dye <vdye@xxxxxxxxxx>
>
> Correct tracking of the 'cache_bottom' for cases where sparse directories
> are present in the index.
Thank you for the detailed background (that I cut from my reply).
> CORRECTING CACHE_BOTTOM
> -----------------------
> The problems observed in 't1092' come from 'cache_bottom' lagging behind in
> cases where the cache tree-based advancement doesn't occur. To solve this,
> then, the fix in 17a1bb570b is "reversed"; rather than skipping
> 'cache_bottom' advancement in 'mark_ce_used()', we skip the directory
> contents-based advancement for sparse directories. Now, every index entry
> can be accounted for in 'cache_bottom':
I have sufficient background to be confident that you are doing the
right thing here!
> -test_expect_failure 'add outside sparse cone' '
> +test_expect_success 'add outside sparse cone' '
> -test_expect_failure 'status/add: outside sparse cone' '
> +test_expect_success 'status/add: outside sparse cone' '
> -test_expect_failure 'reset with pathspecs inside sparse definition' '
> +test_expect_success 'reset with pathspecs inside sparse definition' '
Love to see it.
> if (matches) {
> - o->cache_bottom += matches;
> + /*
> + * Only increment the cache_bottom if the
> + * directory isn't a sparse directory index
> + * entry (if it is, it was already incremented)
> + * in 'mark_ce_used()'
> + */
> + if (!src[0] || !S_ISSPARSEDIR(src[0]->ce_mode))
> + o->cache_bottom += matches;
> return mask;
Looks great.
Thanks,
-Stolee
