Re: [PATCH 2/2] bpf: htab: Reduce elem_size by 8 bytes for small key sizes

[Mykyta Yatsenko <mykyta.yatsenko5@gmail.com>]

On 7/22/26 9:38 PM, T.J. Mercier wrote:
> For standard and PCPU (non-LRU) hash maps with small key sizes (<= 8
> bytes), comparing keys requires only an 8 byte compare. Storing a cached
> 32-bit hash value to shortcut full key comparisons provides no
> performance advantage for small keys, and consumes memory for every
> element.
> 
> This memory can be saved by eliminating hash along with its associated
> 4 byte padding before the key, reducing the elem_size (and key_offset)
> by 8 bytes for standard and PCPU maps.
> 
> Introduce htab_has_hash() to check whether a map requires a cached hash
> field. Update htab_elem_set_hash(), lookup_elem_raw(), and
> lookup_nulls_elem_raw() to conditionally bypass hash checking and
> storage when htab_has_hash() is false.
> 
> Together with the previous patch, this reduces the minimum standard and
> preallocated hash map element size from 64 bytes down to 32 bytes, and
> non-preallocated per-CPU element size from 64 bytes down to 40 bytes.
> 
> Signed-off-by: T.J. Mercier <tjmercier@google.com>
> ---
>  kernel/bpf/hashtab.c                          | 35 +++++++++++++------
>  .../selftests/bpf/progs/map_ptr_kern.c        |  2 +-
>  2 files changed, 25 insertions(+), 12 deletions(-)
> 
> diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
> index 57729c3dff3d..b8c30d402958 100644
> --- a/kernel/bpf/hashtab.c
> +++ b/kernel/bpf/hashtab.c
...
>  
>  	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
> -		if (htab_elem_hash(htab, l) == hash &&
> +		if ((!htab_has_hash(htab) || htab_elem_hash(htab, l) == hash) &&
>  		    !memcmp(htab_elem_key(htab, l), key, key_size))
>  			return l;
>  
> @@ -812,7 +825,7 @@ static struct htab_elem *lookup_nulls_elem_raw(struct bpf_htab *htab,
>  
>  again:
>  	hlist_nulls_for_each_entry_rcu(l, n, head, hash_node)
> -		if (htab_elem_hash(htab, l) == hash &&
> +		if ((!htab_has_hash(htab) || htab_elem_hash(htab, l) == hash) &&

htab_has_hash() is constant for any concrete map, maybe we can store it in the 
struct bpf_htab (it has few holes) and we already describe layout there, 
or at least not calculate on each element in the bucket in the loop.

>  		    !memcmp(htab_elem_key(htab, l), key, key_size))
>  			return l;
>  
> @@ -3219,7 +3232,7 @@ static int rhtab_map_check_btf(struct bpf_map *map, const struct btf *btf,
>  {
>  	struct bpf_rhtab *rhtab = container_of(map, struct bpf_rhtab, map);
>  
> -	return bpf_ma_set_dtor(map, &rhtab->ma, rhtab_mem_dtor);
> +	return bpf_ma_set_dtor(map, &rhtab->ma, rhtab_mem_dtor, offsetof(struct rhtab_elem, data));
>  }
>  
>  static void rhtab_map_free_internal_structs(struct bpf_map *map)
> diff --git a/tools/testing/selftests/bpf/progs/map_ptr_kern.c b/tools/testing/selftests/bpf/progs/map_ptr_kern.c
> index 373c8d17ea55..6bd4cb68c20c 100644
> --- a/tools/testing/selftests/bpf/progs/map_ptr_kern.c
> +++ b/tools/testing/selftests/bpf/progs/map_ptr_kern.c
> @@ -114,7 +114,7 @@ static inline int check_hash(void)
>  	VERIFY(check_default_noinline(&hash->map, map));
>  
>  	VERIFY(hash->n_buckets == MAX_ENTRIES);
> -	VERIFY(hash->elem_size == 64);
> +	VERIFY(hash->elem_size == 32);
>  
>  	VERIFY(hash->count.counter == 0);
>  	VERIFY(bpf_map_sum_elem_count(map) == 0);