-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Pedro Tammela says: ==================== net/sched: retpoline wrappers for tc In tc all qdics, classifiers and actions can be compiled as modules. This results today in indirect calls in all transitions in the tc hierarchy. Due to CONFIG_RETPOLINE, CPUs with mitigations=on might pay an extra cost on indirect calls. For newer Intel cpus with IBRS the extra cost is nonexistent, but AMD Zen cpus and older x86 cpus still go through the retpoline thunk. Known built-in symbols can be optimized into direct calls, thus avoiding the retpoline thunk. So far, tc has not been leveraging this build information and leaving out a performance optimization for some CPUs. In this series we wire up 'tcf_classify()' and 'tcf_action_exec()' with direct calls when known modules are compiled as built-in as an opt-in optimization. We measured these changes in one AMD Zen 4 cpu (Retpoline), one AMD Zen 3 cpu (Retpoline), one Intel 10th Gen CPU (IBRS), one Intel 3rd Gen cpu (Retpoline) and one Intel Xeon CPU (IBRS) using pktgen with 64b udp packets. Our test setup is a dummy device with clsact and matchall in a kernel compiled with every tc module as built-in. We observed a 3-8% speed up on the retpoline CPUs, when going through 1 tc filter, and a 60-100% speed up when going through 100 filters. For the IBRS cpus we observed a 1-2% degradation in both scenarios, we believe the extra branches check introduced a small overhead therefore we added a static key that bypasses the wrapper on kernels not using the retpoline mitigation, but compiled with CONFIG_RETPOLINE. 1 filter: CPU | before (pps) | after (pps) | diff R9 7950X | 5914980 | 6380227 | +7.8% R9 5950X | 4237838 | 4412241 | +4.1% R9 5950X | 4265287 | 4413757 | +3.4% [*] i5-3337U | 1580565 | 1682406 | +6.4% i5-10210U | 3006074 | 3006857 | +0.0% i5-10210U | 3160245 | 3179945 | +0.6% [*] Xeon 6230R | 3196906 | 3197059 | +0.0% Xeon 6230R | 3190392 | 3196153 | +0.01% [*] 100 filters: CPU | before (pps) | after (pps) | diff R9 7950X | 373598 | 820396 | +119.59% R9 5950X | 313469 | 633303 | +102.03% R9 5950X | 313797 | 633150 | +101.77% [*] i5-3337U | 127454 | 211210 | +65.71% i5-10210U | 389259 | 381765 | -1.9% i5-10210U | 408812 | 412730 | +0.9% [*] Xeon 6230R | 415420 | 406612 | -2.1% Xeon 6230R | 416705 | 405869 | -2.6% [*] [*] In these tests we ran pktgen with clone set to 1000. On the 7950x system we also tested the impact of filters if iteration order placement varied, first by compiling a kernel with the filter under test being the first one in the static iteration and then repeating it with being last (of 15 classifiers existing today). We saw a difference of +0.5-1% in pps between being the first in the iteration vs being the last. Therefore we order the classifiers and actions according to relevance per our current thinking. v5->v6: - Address Eric Dumazet suggestions v4->v5: - Rebase v3->v4: - Address Eric Dumazet suggestions v2->v3: - Address suggestions by Jakub, Paolo and Eric - Dropped RFC tag (I forgot to add it on v2) v1->v2: - Fix build errors found by the bots - Address Kuniyuki Iwashima suggestions ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
- Loading branch information
Showing
38 changed files
with
391 additions
and
72 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,251 @@ | ||
/* SPDX-License-Identifier: GPL-2.0 */ | ||
#ifndef __NET_TC_WRAPPER_H | ||
#define __NET_TC_WRAPPER_H | ||
|
||
#include <net/pkt_cls.h> | ||
|
||
#if IS_ENABLED(CONFIG_RETPOLINE) | ||
|
||
#include <linux/cpufeature.h> | ||
#include <linux/static_key.h> | ||
#include <linux/indirect_call_wrapper.h> | ||
|
||
#define TC_INDIRECT_SCOPE | ||
|
||
extern struct static_key_false tc_skip_wrapper; | ||
|
||
/* TC Actions */ | ||
#ifdef CONFIG_NET_CLS_ACT | ||
|
||
#define TC_INDIRECT_ACTION_DECLARE(fname) \ | ||
INDIRECT_CALLABLE_DECLARE(int fname(struct sk_buff *skb, \ | ||
const struct tc_action *a, \ | ||
struct tcf_result *res)) | ||
|
||
TC_INDIRECT_ACTION_DECLARE(tcf_bpf_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_connmark_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_csum_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_ct_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_ctinfo_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_gact_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_gate_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_ife_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_ipt_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_mirred_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_mpls_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_nat_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_pedit_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_police_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_sample_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_simp_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_skbedit_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_skbmod_act); | ||
TC_INDIRECT_ACTION_DECLARE(tcf_vlan_act); | ||
TC_INDIRECT_ACTION_DECLARE(tunnel_key_act); | ||
|
||
static inline int tc_act(struct sk_buff *skb, const struct tc_action *a, | ||
struct tcf_result *res) | ||
{ | ||
if (static_branch_likely(&tc_skip_wrapper)) | ||
goto skip; | ||
|
||
#if IS_BUILTIN(CONFIG_NET_ACT_GACT) | ||
if (a->ops->act == tcf_gact_act) | ||
return tcf_gact_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_MIRRED) | ||
if (a->ops->act == tcf_mirred_act) | ||
return tcf_mirred_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_PEDIT) | ||
if (a->ops->act == tcf_pedit_act) | ||
return tcf_pedit_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_SKBEDIT) | ||
if (a->ops->act == tcf_skbedit_act) | ||
return tcf_skbedit_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_SKBMOD) | ||
if (a->ops->act == tcf_skbmod_act) | ||
return tcf_skbmod_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_POLICE) | ||
if (a->ops->act == tcf_police_act) | ||
return tcf_police_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_BPF) | ||
if (a->ops->act == tcf_bpf_act) | ||
return tcf_bpf_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_CONNMARK) | ||
if (a->ops->act == tcf_connmark_act) | ||
return tcf_connmark_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_CSUM) | ||
if (a->ops->act == tcf_csum_act) | ||
return tcf_csum_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_CT) | ||
if (a->ops->act == tcf_ct_act) | ||
return tcf_ct_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_CTINFO) | ||
if (a->ops->act == tcf_ctinfo_act) | ||
return tcf_ctinfo_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_GATE) | ||
if (a->ops->act == tcf_gate_act) | ||
return tcf_gate_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_MPLS) | ||
if (a->ops->act == tcf_mpls_act) | ||
return tcf_mpls_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_NAT) | ||
if (a->ops->act == tcf_nat_act) | ||
return tcf_nat_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_TUNNEL_KEY) | ||
if (a->ops->act == tunnel_key_act) | ||
return tunnel_key_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_VLAN) | ||
if (a->ops->act == tcf_vlan_act) | ||
return tcf_vlan_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_IFE) | ||
if (a->ops->act == tcf_ife_act) | ||
return tcf_ife_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_IPT) | ||
if (a->ops->act == tcf_ipt_act) | ||
return tcf_ipt_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_SIMP) | ||
if (a->ops->act == tcf_simp_act) | ||
return tcf_simp_act(skb, a, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_ACT_SAMPLE) | ||
if (a->ops->act == tcf_sample_act) | ||
return tcf_sample_act(skb, a, res); | ||
#endif | ||
|
||
skip: | ||
return a->ops->act(skb, a, res); | ||
} | ||
|
||
#endif /* CONFIG_NET_CLS_ACT */ | ||
|
||
/* TC Filters */ | ||
#ifdef CONFIG_NET_CLS | ||
|
||
#define TC_INDIRECT_FILTER_DECLARE(fname) \ | ||
INDIRECT_CALLABLE_DECLARE(int fname(struct sk_buff *skb, \ | ||
const struct tcf_proto *tp, \ | ||
struct tcf_result *res)) | ||
|
||
TC_INDIRECT_FILTER_DECLARE(basic_classify); | ||
TC_INDIRECT_FILTER_DECLARE(cls_bpf_classify); | ||
TC_INDIRECT_FILTER_DECLARE(cls_cgroup_classify); | ||
TC_INDIRECT_FILTER_DECLARE(fl_classify); | ||
TC_INDIRECT_FILTER_DECLARE(flow_classify); | ||
TC_INDIRECT_FILTER_DECLARE(fw_classify); | ||
TC_INDIRECT_FILTER_DECLARE(mall_classify); | ||
TC_INDIRECT_FILTER_DECLARE(route4_classify); | ||
TC_INDIRECT_FILTER_DECLARE(rsvp_classify); | ||
TC_INDIRECT_FILTER_DECLARE(rsvp6_classify); | ||
TC_INDIRECT_FILTER_DECLARE(tcindex_classify); | ||
TC_INDIRECT_FILTER_DECLARE(u32_classify); | ||
|
||
static inline int tc_classify(struct sk_buff *skb, const struct tcf_proto *tp, | ||
struct tcf_result *res) | ||
{ | ||
if (static_branch_likely(&tc_skip_wrapper)) | ||
goto skip; | ||
|
||
#if IS_BUILTIN(CONFIG_NET_CLS_BPF) | ||
if (tp->classify == cls_bpf_classify) | ||
return cls_bpf_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_U32) | ||
if (tp->classify == u32_classify) | ||
return u32_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_FLOWER) | ||
if (tp->classify == fl_classify) | ||
return fl_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_FW) | ||
if (tp->classify == fw_classify) | ||
return fw_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_MATCHALL) | ||
if (tp->classify == mall_classify) | ||
return mall_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_BASIC) | ||
if (tp->classify == basic_classify) | ||
return basic_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_CGROUP) | ||
if (tp->classify == cls_cgroup_classify) | ||
return cls_cgroup_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_FLOW) | ||
if (tp->classify == flow_classify) | ||
return flow_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_ROUTE4) | ||
if (tp->classify == route4_classify) | ||
return route4_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_RSVP) | ||
if (tp->classify == rsvp_classify) | ||
return rsvp_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_RSVP6) | ||
if (tp->classify == rsvp6_classify) | ||
return rsvp6_classify(skb, tp, res); | ||
#endif | ||
#if IS_BUILTIN(CONFIG_NET_CLS_TCINDEX) | ||
if (tp->classify == tcindex_classify) | ||
return tcindex_classify(skb, tp, res); | ||
#endif | ||
|
||
skip: | ||
return tp->classify(skb, tp, res); | ||
} | ||
|
||
static inline void tc_wrapper_init(void) | ||
{ | ||
#ifdef CONFIG_X86 | ||
if (!cpu_feature_enabled(X86_FEATURE_RETPOLINE)) | ||
static_branch_enable(&tc_skip_wrapper); | ||
#endif | ||
} | ||
|
||
#endif /* CONFIG_NET_CLS */ | ||
|
||
#else | ||
|
||
#define TC_INDIRECT_SCOPE static | ||
|
||
static inline int tc_act(struct sk_buff *skb, const struct tc_action *a, | ||
struct tcf_result *res) | ||
{ | ||
return a->ops->act(skb, a, res); | ||
} | ||
|
||
static inline int tc_classify(struct sk_buff *skb, const struct tcf_proto *tp, | ||
struct tcf_result *res) | ||
{ | ||
return tp->classify(skb, tp, res); | ||
} | ||
|
||
static inline void tc_wrapper_init(void) | ||
{ | ||
} | ||
|
||
#endif | ||
|
||
#endif /* __NET_TC_WRAPPER_H */ |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.