Re: [RFC PATCH 05/10] pack-bitmap: don't rely on bitmap_git->reuse_objects

From: Jonathan Tan <jonathantanmy@google.com>
Date: Oct 10, 2019, 23:44 UTC
Message-ID: <20191010234408.170356-1-jonathantanmy@google.com>
In-Reply-To: <20190913130226.7449-6-chriscool@tuxfamily.org>

> As we now allocate 2 more words than necessary for each
> bitmap to serve as marks telling us that we can stop
> iterating over the words, we don't need to rely on
> bitmap_git->reuse_objects to stop iterating over the words.

As Peff states [1], this justification is probably incorrect as well. The actual justification seems to be that we will no longer compute reuse_objects (in a future patch), so we cannot rely on it anymore to terminate the loop early; we have to iterate to the end.

[1] https://public-inbox.org/git/20191002155721.GD6116@sigill.intra.peff.net/

Show 26 quoted lines

> @@ -622,7 +622,7 @@ static void show_objects_for_type(
>  	enum object_type object_type,
>  	show_reachable_fn show_reach)
>  {
> -	size_t pos = 0, i = 0;
> +	size_t i = 0;
>  	uint32_t offset;
>  
>  	struct ewah_iterator it;
> @@ -630,13 +630,15 @@ static void show_objects_for_type(
>  
>  	struct bitmap *objects = bitmap_git->result;
>  
> -	if (bitmap_git->reuse_objects == bitmap_git->pack->num_objects)
> -		return;
> -
>  	ewah_iterator_init(&it, type_filter);
>  
> -	while (i < objects->word_alloc && ewah_iterator_next(&filter, &it)) {
> +	for (i = 0; i < objects->word_alloc &&
> +			ewah_iterator_next(&filter, &it); i++) {
>  		eword_t word = objects->words[i] & filter;
> +		size_t pos = (i * BITS_IN_EWORD);
> +
> +		if (!word)
> +			continue;

Here, iteration is not terminated when we see a 0. We just proceed to the next one.