[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240117095414.GL588419@kernel.org>
Date: Wed, 17 Jan 2024 09:54:14 +0000
From: Simon Horman <horms@...nel.org>
To: Julian Anastasov <ja@....bg>
Cc: lvs-devel@...r.kernel.org, netfilter-devel@...r.kernel.org,
netdev@...r.kernel.org, Dust Li <dust.li@...ux.alibaba.com>,
Jiejian Wu <jiejian@...ux.alibaba.com>,
Jiri Wiesner <jwiesner@...e.de>
Subject: Re: [PATCHv2 RFC net-next 08/14] ipvs: use resizable hash table for
services
On Tue, Dec 12, 2023 at 06:24:38PM +0200, Julian Anastasov wrote:
> Make the hash table for services resizable in the bit range of 4-20.
> Table is attached only while services are present. Resizing is done
> by delayed work based on load (the number of hashed services).
> Table grows when load increases 2+ times (above 12.5% with factor=3)
> and shrinks 8+ times when load decreases 16+ times (below 0.78%).
>
> Switch to jhash hashing to reduce the collisions for multiple
> services.
>
> Add a hash_key field into the service that includes table ID and
> bucket ID. This helps the lookup and delete operations.
>
> Signed-off-by: Julian Anastasov <ja@....bg>
...
> @@ -391,18 +440,29 @@ static inline struct ip_vs_service *
> __ip_vs_service_find(struct netns_ipvs *ipvs, int af, __u16 protocol,
> const union nf_inet_addr *vaddr, __be16 vport)
> {
> - unsigned int hash;
> + DECLARE_IP_VS_RHT_WALK_BUCKET_RCU();
> + struct hlist_bl_head *head;
> struct ip_vs_service *svc;
> -
> - /* Check for "full" addressed entries */
> - hash = ip_vs_svc_hashkey(ipvs, af, protocol, vaddr, vport);
> -
> - hlist_for_each_entry_rcu(svc, &ipvs->svc_table[hash], s_list) {
> - if (svc->af == af && ip_vs_addr_equal(af, &svc->addr, vaddr) &&
> - svc->port == vport && svc->protocol == protocol &&
> - !svc->fwmark) {
> - /* HIT */
> - return svc;
> + struct ip_vs_rht *t, *p;
> + struct hlist_bl_node *e;
> + u32 hash, hash_key;
> +
> + ip_vs_rht_for_each_table_rcu(ipvs->svc_table, t, p) {
> + /* Check for "full" addressed entries */
> + hash = ip_vs_svc_hashval(t, af, protocol, vaddr, vport);
> +
> + hash_key = ip_vs_rht_build_hash_key(t, hash);
> + ip_vs_rht_walk_bucket_rcu(t, hash_key, head) {
> + hlist_bl_for_each_entry_rcu(svc, e, head, s_list) {
Hi Julian,
Smatch thinks that head is used uninitialised here.
This does seem to be the case to me too.
> + if (READ_ONCE(svc->hash_key) == hash_key &&
> + svc->af == af &&
> + ip_vs_addr_equal(af, &svc->addr, vaddr) &&
> + svc->port == vport &&
> + svc->protocol == protocol && !svc->fwmark) {
> + /* HIT */
> + return svc;
> + }
> + }
> }
> }
>
> @@ -416,16 +476,26 @@ __ip_vs_service_find(struct netns_ipvs *ipvs, int af, __u16 protocol,
> static inline struct ip_vs_service *
> __ip_vs_svc_fwm_find(struct netns_ipvs *ipvs, int af, __u32 fwmark)
> {
> - unsigned int hash;
> + DECLARE_IP_VS_RHT_WALK_BUCKET_RCU();
> + struct hlist_bl_head *head;
> struct ip_vs_service *svc;
> -
> - /* Check for fwmark addressed entries */
> - hash = ip_vs_svc_fwm_hashkey(ipvs, fwmark);
> -
> - hlist_for_each_entry_rcu(svc, &ipvs->svc_table[hash], s_list) {
> - if (svc->fwmark == fwmark && svc->af == af) {
> - /* HIT */
> - return svc;
> + struct ip_vs_rht *t, *p;
> + struct hlist_bl_node *e;
> + u32 hash, hash_key;
> +
> + ip_vs_rht_for_each_table_rcu(ipvs->svc_table, t, p) {
> + /* Check for fwmark addressed entries */
> + hash = ip_vs_svc_fwm_hashval(t, af, fwmark);
> +
> + hash_key = ip_vs_rht_build_hash_key(t, hash);
> + ip_vs_rht_walk_bucket_rcu(t, hash_key, head) {
> + hlist_bl_for_each_entry_rcu(svc, e, head, s_list) {
Ditto.
> + if (READ_ONCE(svc->hash_key) == hash_key &&
> + svc->fwmark == fwmark && svc->af == af) {
> + /* HIT */
> + return svc;
> + }
> + }
> }
> }
>
...
Powered by blists - more mailing lists