Re: [PATCH v2 0/5] Skip 'cache_tree_update()' when 'prime_cache_tree()' is called immediate after

From: Taylor Blau <me@ttaylorr.com>
Date: Nov 10, 2022, 02:12 UTC
Message-ID: <Y2xeJmkMVnn0tk5V@nand.local>
In-Reply-To: <pull.1411.v2.git.1668045438.gitgitgadget@gmail.com>

On Thu, Nov 10, 2022 at 01:57:12AM +0000, Victoria Dye via GitGitGadget wrote:

Show 12 quoted lines

> Changes since V1
> ================
>
>  * Rewrote 'p0090' to more accurately and reliably test 'prime_cache_tree()'
>    vs. 'cache_tree_update()'.
>    * Moved iterative cache tree update out of C and into the shell tests (to
>      avoid potential runtime optimizations)
>    * Added a "control" test to document how much of the execution time is
>      startup overhead
>    * Added tests demonstrating performance in partially-invalid cache trees.
>  * Fixed the use of 'prime_cache_tree()' in 'test-tool cache-tree', changing
>    it from using the tree at HEAD to the current cache tree.

All seem very reasonable to me, and the range-diff matches what you say.

Let's hear from Stolee, who reviewed the first round, too, and then we should feel comfortable to start merging this down.

Thanks, Taylor