Commit 0a7960c7 authored by David S. Miller's avatar David S. Miller

Merge branch 'net-sched-Add-txtime-assist-support-for-taprio'

Vedang Patel says:

====================
net/sched: Add txtime-assist support for taprio.

Changes in v6:
- Use _BITUL() instead of BIT() in UAPI for etf. (patch #1)
- Fix a bug reported by kbuild test bot in length_to_duration(). (patch #6)
- Remove an unused function (get_cycle_start()). (Patch #6)

Changes in v5:
- Commit message improved for the igb patch (patch #1).
- Fixed typo in commit message for etf patch (patch #2).

Changes in v4:
- Remove inline directive from functions in foo.c.
- Fix spacing in pkt_sched.h (for etf patch).

Changes in v3:
- Simplify implementation for taprio flags.
- txtime_delay can only be set if txtime-assist mode is enabled.
- txtime_delay and flags will only be visible in tc output if set by user.
- Minor changes in error reporting.

Changes in v2:
- Txtime-offload has now been renamed to txtime-assist mode.
- Renamed the offload parameter to flags.
- Removed the code which introduced the hardware offloading functionality.

Original Cover letter (with above changes included)
--------------------------------------------------

Currently, we are seeing packets being transmitted outside their
timeslices. We can confirm that the packets are being dequeued at the right
time. So, the delay is induced after the packet is dequeued, because
taprio, without any offloading, has no control of when a packet is actually
transmitted.

In order to solve this, we are making use of the txtime feature provided by
ETF qdisc. Hardware offloading needs to be supported by the ETF qdisc in
order to take advantage of this feature. The taprio qdisc will assign
txtime (in skb->tstamp) for all the packets which do not have the txtime
allocated via the SO_TXTIME socket option. For the packets which already
have SO_TXTIME set, taprio will validate whether the packet will be
transmitted in the correct interval.

In order to support this, the following parameters have been added:
- flags (taprio): This is added in order to support different offloading
  modes which will be added in the future.
- txtime-delay (taprio): This indicates the minimum time it will take for
  the packet to hit the wire after it reaches taprio_enqueue(). This is
  useful in determining whether we can transmit the packet in the remaining
  time if the gate corresponding to the packet is currently open.
- skip_skb_check (ETF): ETF currently drops any packet which does not have
  the SO_TXTIME socket option set. This check can be skipped by specifying
  this option.

Following is an example configuration:

tc qdisc replace dev $IFACE parent root handle 100 taprio \\
    num_tc 3 \\
    map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 \\
    queues 1@0 1@0 1@0 \\
    base-time $BASE_TIME \\
    sched-entry S 01 300000 \\
    sched-entry S 02 300000 \\
    sched-entry S 04 400000 \\
    flags 0x1 \\
    txtime-delay 200000 \\
    clockid CLOCK_TAI

tc qdisc replace dev $IFACE parent 100:1 etf \\
    offload delta 200000 clockid CLOCK_TAI skip_skb_check

Here, the "flags" parameter is indicating that the txtime-assist mode is
enabled. Also, all the traffic classes have been assigned the same queue.
This is to prevent the traffic classes in the lower priority queues from
getting starved. Note that this configuration is specific to the i210
ethernet card. Other network cards where the hardware queues are given the
same priority, might be able to utilize more than one queue.

Following are some of the other highlights of the series:
- Fix a bug where hardware timestamping and SO_TXTIME options cannot be
  used together. (Patch 1)
- Introduces the skip_skb_check option.  (Patch 2)
- Make TxTime assist mode work with TCP packets (Patch 7).

The following changes are recommended to be done in order to get the best
performance from taprio in this mode:
ip link set dev enp1s0 mtu 1514
ethtool -K eth0 gso off
ethtool -K eth0 tso off
ethtool --set-eee eth0 eee off
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents 8747d82d 54002066
...@@ -5688,6 +5688,7 @@ static void igb_tx_ctxtdesc(struct igb_ring *tx_ring, ...@@ -5688,6 +5688,7 @@ static void igb_tx_ctxtdesc(struct igb_ring *tx_ring,
*/ */
if (tx_ring->launchtime_enable) { if (tx_ring->launchtime_enable) {
ts = ns_to_timespec64(first->skb->tstamp); ts = ns_to_timespec64(first->skb->tstamp);
first->skb->tstamp = 0;
context_desc->seqnum_seed = cpu_to_le32(ts.tv_nsec / 32); context_desc->seqnum_seed = cpu_to_le32(ts.tv_nsec / 32);
} else { } else {
context_desc->seqnum_seed = 0; context_desc->seqnum_seed = 0;
......
...@@ -988,8 +988,9 @@ struct tc_etf_qopt { ...@@ -988,8 +988,9 @@ struct tc_etf_qopt {
__s32 delta; __s32 delta;
__s32 clockid; __s32 clockid;
__u32 flags; __u32 flags;
#define TC_ETF_DEADLINE_MODE_ON BIT(0) #define TC_ETF_DEADLINE_MODE_ON _BITUL(0)
#define TC_ETF_OFFLOAD_ON BIT(1) #define TC_ETF_OFFLOAD_ON _BITUL(1)
#define TC_ETF_SKIP_SOCK_CHECK _BITUL(2)
}; };
enum { enum {
...@@ -1158,6 +1159,8 @@ enum { ...@@ -1158,6 +1159,8 @@ enum {
* [TCA_TAPRIO_ATTR_SCHED_ENTRY_INTERVAL] * [TCA_TAPRIO_ATTR_SCHED_ENTRY_INTERVAL]
*/ */
#define TCA_TAPRIO_ATTR_FLAG_TXTIME_ASSIST 0x1
enum { enum {
TCA_TAPRIO_ATTR_UNSPEC, TCA_TAPRIO_ATTR_UNSPEC,
TCA_TAPRIO_ATTR_PRIOMAP, /* struct tc_mqprio_qopt */ TCA_TAPRIO_ATTR_PRIOMAP, /* struct tc_mqprio_qopt */
...@@ -1169,6 +1172,8 @@ enum { ...@@ -1169,6 +1172,8 @@ enum {
TCA_TAPRIO_ATTR_ADMIN_SCHED, /* The admin sched, only used in dump */ TCA_TAPRIO_ATTR_ADMIN_SCHED, /* The admin sched, only used in dump */
TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME, /* s64 */ TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME, /* s64 */
TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME_EXTENSION, /* s64 */ TCA_TAPRIO_ATTR_SCHED_CYCLE_TIME_EXTENSION, /* s64 */
TCA_TAPRIO_ATTR_FLAGS, /* u32 */
TCA_TAPRIO_ATTR_TXTIME_DELAY, /* s32 */
__TCA_TAPRIO_ATTR_MAX, __TCA_TAPRIO_ATTR_MAX,
}; };
......
...@@ -22,10 +22,12 @@ ...@@ -22,10 +22,12 @@
#define DEADLINE_MODE_IS_ON(x) ((x)->flags & TC_ETF_DEADLINE_MODE_ON) #define DEADLINE_MODE_IS_ON(x) ((x)->flags & TC_ETF_DEADLINE_MODE_ON)
#define OFFLOAD_IS_ON(x) ((x)->flags & TC_ETF_OFFLOAD_ON) #define OFFLOAD_IS_ON(x) ((x)->flags & TC_ETF_OFFLOAD_ON)
#define SKIP_SOCK_CHECK_IS_SET(x) ((x)->flags & TC_ETF_SKIP_SOCK_CHECK)
struct etf_sched_data { struct etf_sched_data {
bool offload; bool offload;
bool deadline_mode; bool deadline_mode;
bool skip_sock_check;
int clockid; int clockid;
int queue; int queue;
s32 delta; /* in ns */ s32 delta; /* in ns */
...@@ -77,6 +79,9 @@ static bool is_packet_valid(struct Qdisc *sch, struct sk_buff *nskb) ...@@ -77,6 +79,9 @@ static bool is_packet_valid(struct Qdisc *sch, struct sk_buff *nskb)
struct sock *sk = nskb->sk; struct sock *sk = nskb->sk;
ktime_t now; ktime_t now;
if (q->skip_sock_check)
goto skip;
if (!sk) if (!sk)
return false; return false;
...@@ -92,6 +97,7 @@ static bool is_packet_valid(struct Qdisc *sch, struct sk_buff *nskb) ...@@ -92,6 +97,7 @@ static bool is_packet_valid(struct Qdisc *sch, struct sk_buff *nskb)
if (sk->sk_txtime_deadline_mode != q->deadline_mode) if (sk->sk_txtime_deadline_mode != q->deadline_mode)
return false; return false;
skip:
now = q->get_time(); now = q->get_time();
if (ktime_before(txtime, now) || ktime_before(txtime, q->last)) if (ktime_before(txtime, now) || ktime_before(txtime, q->last))
return false; return false;
...@@ -385,6 +391,7 @@ static int etf_init(struct Qdisc *sch, struct nlattr *opt, ...@@ -385,6 +391,7 @@ static int etf_init(struct Qdisc *sch, struct nlattr *opt,
q->clockid = qopt->clockid; q->clockid = qopt->clockid;
q->offload = OFFLOAD_IS_ON(qopt); q->offload = OFFLOAD_IS_ON(qopt);
q->deadline_mode = DEADLINE_MODE_IS_ON(qopt); q->deadline_mode = DEADLINE_MODE_IS_ON(qopt);
q->skip_sock_check = SKIP_SOCK_CHECK_IS_SET(qopt);
switch (q->clockid) { switch (q->clockid) {
case CLOCK_REALTIME: case CLOCK_REALTIME:
...@@ -473,6 +480,9 @@ static int etf_dump(struct Qdisc *sch, struct sk_buff *skb) ...@@ -473,6 +480,9 @@ static int etf_dump(struct Qdisc *sch, struct sk_buff *skb)
if (q->deadline_mode) if (q->deadline_mode)
opt.flags |= TC_ETF_DEADLINE_MODE_ON; opt.flags |= TC_ETF_DEADLINE_MODE_ON;
if (q->skip_sock_check)
opt.flags |= TC_ETF_SKIP_SOCK_CHECK;
if (nla_put(skb, TCA_ETF_PARMS, sizeof(opt), &opt)) if (nla_put(skb, TCA_ETF_PARMS, sizeof(opt), &opt))
goto nla_put_failure; goto nla_put_failure;
......
...@@ -21,12 +21,17 @@ ...@@ -21,12 +21,17 @@
#include <net/pkt_sched.h> #include <net/pkt_sched.h>
#include <net/pkt_cls.h> #include <net/pkt_cls.h>
#include <net/sch_generic.h> #include <net/sch_generic.h>
#include <net/sock.h>
#include <net/tcp.h>
static LIST_HEAD(taprio_list); static LIST_HEAD(taprio_list);
static DEFINE_SPINLOCK(taprio_list_lock); static DEFINE_SPINLOCK(taprio_list_lock);
#define TAPRIO_ALL_GATES_OPEN -1 #define TAPRIO_ALL_GATES_OPEN -1
#define FLAGS_VALID(flags) (!((flags) & ~TCA_TAPRIO_ATTR_FLAG_TXTIME_ASSIST))
#define TXTIME_ASSIST_IS_ENABLED(flags) ((flags) & TCA_TAPRIO_ATTR_FLAG_TXTIME_ASSIST)
struct sched_entry { struct sched_entry {
struct list_head list; struct list_head list;
...@@ -35,6 +40,7 @@ struct sched_entry { ...@@ -35,6 +40,7 @@ struct sched_entry {
* packet leaves after this time. * packet leaves after this time.
*/ */
ktime_t close_time; ktime_t close_time;
ktime_t next_txtime;
atomic_t budget; atomic_t budget;
int index; int index;
u32 gate_mask; u32 gate_mask;
...@@ -55,6 +61,8 @@ struct sched_gate_list { ...@@ -55,6 +61,8 @@ struct sched_gate_list {
struct taprio_sched { struct taprio_sched {
struct Qdisc **qdiscs; struct Qdisc **qdiscs;
struct Qdisc *root; struct Qdisc *root;
u32 flags;
enum tk_offsets tk_offset;
int clockid; int clockid;
atomic64_t picos_per_byte; /* Using picoseconds because for 10Gbps+ atomic64_t picos_per_byte; /* Using picoseconds because for 10Gbps+
* speeds it's sub-nanoseconds per byte * speeds it's sub-nanoseconds per byte
...@@ -65,9 +73,9 @@ struct taprio_sched { ...@@ -65,9 +73,9 @@ struct taprio_sched {
struct sched_entry __rcu *current_entry; struct sched_entry __rcu *current_entry;
struct sched_gate_list __rcu *oper_sched; struct sched_gate_list __rcu *oper_sched;
struct sched_gate_list __rcu *admin_sched; struct sched_gate_list __rcu *admin_sched;
ktime_t (*get_time)(void);
struct hrtimer advance_timer; struct hrtimer advance_timer;
struct list_head taprio_list; struct list_head taprio_list;
int txtime_delay;
}; };
static ktime_t sched_base_time(const struct sched_gate_list *sched) static ktime_t sched_base_time(const struct sched_gate_list *sched)
...@@ -78,6 +86,20 @@ static ktime_t sched_base_time(const struct sched_gate_list *sched) ...@@ -78,6 +86,20 @@ static ktime_t sched_base_time(const struct sched_gate_list *sched)
return ns_to_ktime(sched->base_time); return ns_to_ktime(sched->base_time);
} }
static ktime_t taprio_get_time(struct taprio_sched *q)
{
ktime_t mono = ktime_get();
switch (q->tk_offset) {
case TK_OFFS_MAX:
return mono;
default:
return ktime_mono_to_any(mono, q->tk_offset);
}
return KTIME_MAX;
}
static void taprio_free_sched_cb(struct rcu_head *head) static void taprio_free_sched_cb(struct rcu_head *head)
{ {
struct sched_gate_list *sched = container_of(head, struct sched_gate_list, rcu); struct sched_gate_list *sched = container_of(head, struct sched_gate_list, rcu);
...@@ -108,20 +130,263 @@ static void switch_schedules(struct taprio_sched *q, ...@@ -108,20 +130,263 @@ static void switch_schedules(struct taprio_sched *q,
*admin = NULL; *admin = NULL;
} }
static ktime_t get_cycle_time(struct sched_gate_list *sched) /* Get how much time has been already elapsed in the current cycle. */
static s32 get_cycle_time_elapsed(struct sched_gate_list *sched, ktime_t time)
{
ktime_t time_since_sched_start;
s32 time_elapsed;
time_since_sched_start = ktime_sub(time, sched->base_time);
div_s64_rem(time_since_sched_start, sched->cycle_time, &time_elapsed);
return time_elapsed;
}
static ktime_t get_interval_end_time(struct sched_gate_list *sched,
struct sched_gate_list *admin,
struct sched_entry *entry,
ktime_t intv_start)
{
s32 cycle_elapsed = get_cycle_time_elapsed(sched, intv_start);
ktime_t intv_end, cycle_ext_end, cycle_end;
cycle_end = ktime_add_ns(intv_start, sched->cycle_time - cycle_elapsed);
intv_end = ktime_add_ns(intv_start, entry->interval);
cycle_ext_end = ktime_add(cycle_end, sched->cycle_time_extension);
if (ktime_before(intv_end, cycle_end))
return intv_end;
else if (admin && admin != sched &&
ktime_after(admin->base_time, cycle_end) &&
ktime_before(admin->base_time, cycle_ext_end))
return admin->base_time;
else
return cycle_end;
}
static int length_to_duration(struct taprio_sched *q, int len)
{
return div_u64(len * atomic64_read(&q->picos_per_byte), 1000);
}
/* Returns the entry corresponding to next available interval. If
* validate_interval is set, it only validates whether the timestamp occurs
* when the gate corresponding to the skb's traffic class is open.
*/
static struct sched_entry *find_entry_to_transmit(struct sk_buff *skb,
struct Qdisc *sch,
struct sched_gate_list *sched,
struct sched_gate_list *admin,
ktime_t time,
ktime_t *interval_start,
ktime_t *interval_end,
bool validate_interval)
{
ktime_t curr_intv_start, curr_intv_end, cycle_end, packet_transmit_time;
ktime_t earliest_txtime = KTIME_MAX, txtime, cycle, transmit_end_time;
struct sched_entry *entry = NULL, *entry_found = NULL;
struct taprio_sched *q = qdisc_priv(sch);
struct net_device *dev = qdisc_dev(sch);
bool entry_available = false;
s32 cycle_elapsed;
int tc, n;
tc = netdev_get_prio_tc_map(dev, skb->priority);
packet_transmit_time = length_to_duration(q, qdisc_pkt_len(skb));
*interval_start = 0;
*interval_end = 0;
if (!sched)
return NULL;
cycle = sched->cycle_time;
cycle_elapsed = get_cycle_time_elapsed(sched, time);
curr_intv_end = ktime_sub_ns(time, cycle_elapsed);
cycle_end = ktime_add_ns(curr_intv_end, cycle);
list_for_each_entry(entry, &sched->entries, list) {
curr_intv_start = curr_intv_end;
curr_intv_end = get_interval_end_time(sched, admin, entry,
curr_intv_start);
if (ktime_after(curr_intv_start, cycle_end))
break;
if (!(entry->gate_mask & BIT(tc)) ||
packet_transmit_time > entry->interval)
continue;
txtime = entry->next_txtime;
if (ktime_before(txtime, time) || validate_interval) {
transmit_end_time = ktime_add_ns(time, packet_transmit_time);
if ((ktime_before(curr_intv_start, time) &&
ktime_before(transmit_end_time, curr_intv_end)) ||
(ktime_after(curr_intv_start, time) && !validate_interval)) {
entry_found = entry;
*interval_start = curr_intv_start;
*interval_end = curr_intv_end;
break;
} else if (!entry_available && !validate_interval) {
/* Here, we are just trying to find out the
* first available interval in the next cycle.
*/
entry_available = 1;
entry_found = entry;
*interval_start = ktime_add_ns(curr_intv_start, cycle);
*interval_end = ktime_add_ns(curr_intv_end, cycle);
}
} else if (ktime_before(txtime, earliest_txtime) &&
!entry_available) {
earliest_txtime = txtime;
entry_found = entry;
n = div_s64(ktime_sub(txtime, curr_intv_start), cycle);
*interval_start = ktime_add(curr_intv_start, n * cycle);
*interval_end = ktime_add(curr_intv_end, n * cycle);
}
}
return entry_found;
}
static bool is_valid_interval(struct sk_buff *skb, struct Qdisc *sch)
{ {
struct taprio_sched *q = qdisc_priv(sch);
struct sched_gate_list *sched, *admin;
ktime_t interval_start, interval_end;
struct sched_entry *entry; struct sched_entry *entry;
ktime_t cycle = 0;
if (sched->cycle_time != 0) rcu_read_lock();
return sched->cycle_time; sched = rcu_dereference(q->oper_sched);
admin = rcu_dereference(q->admin_sched);
entry = find_entry_to_transmit(skb, sch, sched, admin, skb->tstamp,
&interval_start, &interval_end, true);
rcu_read_unlock();
list_for_each_entry(entry, &sched->entries, list) return entry;
cycle = ktime_add_ns(cycle, entry->interval); }
sched->cycle_time = cycle; /* This returns the tstamp value set by TCP in terms of the set clock. */
static ktime_t get_tcp_tstamp(struct taprio_sched *q, struct sk_buff *skb)
{
unsigned int offset = skb_network_offset(skb);
const struct ipv6hdr *ipv6h;
const struct iphdr *iph;
struct ipv6hdr _ipv6h;
return cycle; ipv6h = skb_header_pointer(skb, offset, sizeof(_ipv6h), &_ipv6h);
if (!ipv6h)
return 0;
if (ipv6h->version == 4) {
iph = (struct iphdr *)ipv6h;
offset += iph->ihl * 4;
/* special-case 6in4 tunnelling, as that is a common way to get
* v6 connectivity in the home
*/
if (iph->protocol == IPPROTO_IPV6) {
ipv6h = skb_header_pointer(skb, offset,
sizeof(_ipv6h), &_ipv6h);
if (!ipv6h || ipv6h->nexthdr != IPPROTO_TCP)
return 0;
} else if (iph->protocol != IPPROTO_TCP) {
return 0;
}
} else if (ipv6h->version == 6 && ipv6h->nexthdr != IPPROTO_TCP) {
return 0;
}
return ktime_mono_to_any(skb->skb_mstamp_ns, q->tk_offset);
}
/* There are a few scenarios where we will have to modify the txtime from
* what is read from next_txtime in sched_entry. They are:
* 1. If txtime is in the past,
* a. The gate for the traffic class is currently open and packet can be
* transmitted before it closes, schedule the packet right away.
* b. If the gate corresponding to the traffic class is going to open later
* in the cycle, set the txtime of packet to the interval start.
* 2. If txtime is in the future, there are packets corresponding to the
* current traffic class waiting to be transmitted. So, the following
* possibilities exist:
* a. We can transmit the packet before the window containing the txtime
* closes.
* b. The window might close before the transmission can be completed
* successfully. So, schedule the packet in the next open window.
*/
static long get_packet_txtime(struct sk_buff *skb, struct Qdisc *sch)
{
ktime_t transmit_end_time, interval_end, interval_start, tcp_tstamp;
struct taprio_sched *q = qdisc_priv(sch);
struct sched_gate_list *sched, *admin;
ktime_t minimum_time, now, txtime;
int len, packet_transmit_time;
struct sched_entry *entry;
bool sched_changed;
now = taprio_get_time(q);
minimum_time = ktime_add_ns(now, q->txtime_delay);
tcp_tstamp = get_tcp_tstamp(q, skb);
minimum_time = max_t(ktime_t, minimum_time, tcp_tstamp);
rcu_read_lock();
admin = rcu_dereference(q->admin_sched);
sched = rcu_dereference(q->oper_sched);
if (admin && ktime_after(minimum_time, admin->base_time))
switch_schedules(q, &admin, &sched);
/* Until the schedule starts, all the queues are open */
if (!sched || ktime_before(minimum_time, sched->base_time)) {
txtime = minimum_time;
goto done;
}
len = qdisc_pkt_len(skb);
packet_transmit_time = length_to_duration(q, len);
do {
sched_changed = 0;
entry = find_entry_to_transmit(skb, sch, sched, admin,
minimum_time,
&interval_start, &interval_end,
false);
if (!entry) {
txtime = 0;
goto done;
}
txtime = entry->next_txtime;
txtime = max_t(ktime_t, txtime, minimum_time);
txtime = max_t(ktime_t, txtime, interval_start);
if (admin && admin != sched &&
ktime_after(txtime, admin->base_time)) {
sched = admin;
sched_changed = 1;
continue;
}
transmit_end_time = ktime_add(txtime, packet_transmit_time);
minimum_time = transmit_end_time;
/* Update the txtime of current entry to the next time it's
* interval starts.
*/
if (ktime_after(transmit_end_time, interval_end))
entry->next_txtime = ktime_add(interval_start, sched->cycle_time);
} while (sched_changed || ktime_after(transmit_end_time, interval_end));
entry->next_txtime = transmit_end_time;
done:
rcu_read_unlock();
return txtime;
} }
static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch, static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch,
...@@ -137,6 +402,15 @@ static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch, ...@@ -137,6 +402,15 @@ static int taprio_enqueue(struct sk_buff *skb, struct Qdisc *sch,
if (unlikely(!child)) if (unlikely(!child))
return qdisc_drop(skb, sch, to_free); return qdisc_drop(skb, sch, to_free);
if (skb->sk && sock_flag(skb->sk, SOCK_TXTIME)) {
if (!is_valid_interval(skb, sch))
return qdisc_drop(skb, sch, to_free);
} else if (TXTIME_ASSIST_IS_ENABLED(q->flags)) {
skb->tstamp = get_packet_txtime(skb, sch);
if (!skb->tstamp)
return qdisc_drop(skb, sch, to_free);
}
qdisc_qstats_backlog_inc(sch, skb); qdisc_qstats_backlog_inc(sch, skb);
sch->q.qlen++; sch->q.qlen++;
...@@ -172,6 +446,9 @@ static struct sk_buff *taprio_peek(struct Qdisc *sch) ...@@ -172,6 +446,9 @@ static struct sk_buff *taprio_peek(struct Qdisc *sch)
if (!skb) if (!skb)
continue; continue;
if (TXTIME_ASSIST_IS_ENABLED(q->flags))
return skb;
prio = skb->priority; prio = skb->priority;
tc = netdev_get_prio_tc_map(dev, prio); tc = netdev_get_prio_tc_map(dev, prio);
...@@ -184,11 +461,6 @@ static struct sk_buff *taprio_peek(struct Qdisc *sch) ...@@ -184,11 +461,6 @@ static struct sk_buff *taprio_peek(struct Qdisc *sch)
return NULL; return NULL;
} }
static inline int length_to_duration(struct taprio_sched *q, int len)
{
return div_u64(len * atomic64_read(&q->picos_per_byte), 1000);
}
static void taprio_set_budget(struct taprio_sched *q, struct sched_entry *entry) static void taprio_set_budget(struct taprio_sched *q, struct sched_entry *entry)
{ {
atomic_set(&entry->budget, atomic_set(&entry->budget,
...@@ -232,6 +504,13 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch) ...@@ -232,6 +504,13 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch)
if (unlikely(!child)) if (unlikely(!child))
continue; continue;
if (TXTIME_ASSIST_IS_ENABLED(q->flags)) {
skb = child->ops->dequeue(child);
if (!skb)
continue;
goto skb_found;
}
skb = child->ops->peek(child); skb = child->ops->peek(child);
if (!skb) if (!skb)
continue; continue;
...@@ -243,7 +522,7 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch) ...@@ -243,7 +522,7 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch)
continue; continue;
len = qdisc_pkt_len(skb); len = qdisc_pkt_len(skb);
guard = ktime_add_ns(q->get_time(), guard = ktime_add_ns(taprio_get_time(q),
length_to_duration(q, len)); length_to_duration(q, len));
/* In the case that there's no gate entry, there's no /* In the case that there's no gate entry, there's no
...@@ -262,6 +541,7 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch) ...@@ -262,6 +541,7 @@ static struct sk_buff *taprio_dequeue(struct Qdisc *sch)
if (unlikely(!skb)) if (unlikely(!skb))
goto done; goto done;
skb_found:
qdisc_bstats_update(sch, skb); qdisc_bstats_update(sch, skb);
qdisc_qstats_backlog_dec(sch, skb); qdisc_qstats_backlog_dec(sch, skb);
sch->q.qlen--; sch->q.qlen--;
...@@ -524,12 +804,22 @@ static int parse_taprio_schedule(struct nlattr **tb, ...@@ -524,12 +804,22 @@ static int parse_taprio_schedule(struct nlattr **tb,
if (err < 0) if (err < 0)
return err; return err;
if (!new->cycle_time) {
struct sched_entry *entry;
ktime_t cycle = 0;
list_for_each_entry(entry, &new->entries, list)
cycle = ktime_add_ns(cycle, entry->interval);
new->cycle_time = cycle;
}
return 0; return 0;
} }
static int taprio_parse_mqprio_opt(struct net_device *dev, static int taprio_parse_mqprio_opt(struct net_device *dev,
struct tc_mqprio_qopt *qopt, struct tc_mqprio_qopt *qopt,
struct netlink_ext_ack *extack) struct netlink_ext_ack *extack,
u32 taprio_flags)
{ {
int i, j; int i, j;
...@@ -577,6 +867,9 @@ static int taprio_parse_mqprio_opt(struct net_device *dev, ...@@ -577,6 +867,9 @@ static int taprio_parse_mqprio_opt(struct net_device *dev,
return -EINVAL; return -EINVAL;
} }
if (TXTIME_ASSIST_IS_ENABLED(taprio_flags))
continue;
/* Verify that the offset and counts do not overlap */ /* Verify that the offset and counts do not overlap */
for (j = i + 1; j < qopt->num_tc; j++) { for (j = i + 1; j < qopt->num_tc; j++) {
if (last > qopt->offset[j]) { if (last > qopt->offset[j]) {
...@@ -598,14 +891,14 @@ static int taprio_get_start_time(struct Qdisc *sch, ...@@ -598,14 +891,14 @@ static int taprio_get_start_time(struct Qdisc *sch,
s64 n; s64 n;
base = sched_base_time(sched); base = sched_base_time(sched);
now = q->get_time(); now = taprio_get_time(q);
if (ktime_after(base, now)) { if (ktime_after(base, now)) {
*start = base; *start = base;
return 0; return 0;
} }
cycle = get_cycle_time(sched); cycle = sched->cycle_time;
/* The qdisc is expected to have at least one sched_entry. Moreover, /* The qdisc is expected to have at least one sched_entry. Moreover,
* any entry must have 'interval' > 0. Thus if the cycle time is zero, * any entry must have 'interval' > 0. Thus if the cycle time is zero,
...@@ -632,7 +925,7 @@ static void setup_first_close_time(struct taprio_sched *q, ...@@ -632,7 +925,7 @@ static void setup_first_close_time(struct taprio_sched *q,
first = list_first_entry(&sched->entries, first = list_first_entry(&sched->entries,
struct sched_entry, list); struct sched_entry, list);
cycle = get_cycle_time(sched); cycle = sched->cycle_time;
/* FIXME: find a better place to do this */ /* FIXME: find a better place to do this */
sched->cycle_close_time = ktime_add_ns(base, cycle); sched->cycle_close_time = ktime_add_ns(base, cycle);
...@@ -707,6 +1000,18 @@ static int taprio_dev_notifier(struct notifier_block *nb, unsigned long event, ...@@ -707,6 +1000,18 @@ static int taprio_dev_notifier(struct notifier_block *nb, unsigned long event,
return NOTIFY_DONE; return NOTIFY_DONE;
} }
static void setup_txtime(struct taprio_sched *q,
struct sched_gate_list *sched, ktime_t base)
{
struct sched_entry *entry;
u32 interval = 0;
list_for_each_entry(entry, &sched->entries, list) {
entry->next_txtime = ktime_add_ns(base, interval);
interval += entry->interval;
}
}
static int taprio_change(struct Qdisc *sch, struct nlattr *opt, static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
struct netlink_ext_ack *extack) struct netlink_ext_ack *extack)
{ {
...@@ -715,6 +1020,7 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt, ...@@ -715,6 +1020,7 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
struct taprio_sched *q = qdisc_priv(sch); struct taprio_sched *q = qdisc_priv(sch);
struct net_device *dev = qdisc_dev(sch); struct net_device *dev = qdisc_dev(sch);
struct tc_mqprio_qopt *mqprio = NULL; struct tc_mqprio_qopt *mqprio = NULL;
u32 taprio_flags = 0;
int i, err, clockid; int i, err, clockid;
unsigned long flags; unsigned long flags;
ktime_t start; ktime_t start;
...@@ -727,7 +1033,21 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt, ...@@ -727,7 +1033,21 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
if (tb[TCA_TAPRIO_ATTR_PRIOMAP]) if (tb[TCA_TAPRIO_ATTR_PRIOMAP])
mqprio = nla_data(tb[TCA_TAPRIO_ATTR_PRIOMAP]); mqprio = nla_data(tb[TCA_TAPRIO_ATTR_PRIOMAP]);
err = taprio_parse_mqprio_opt(dev, mqprio, extack); if (tb[TCA_TAPRIO_ATTR_FLAGS]) {
taprio_flags = nla_get_u32(tb[TCA_TAPRIO_ATTR_FLAGS]);
if (q->flags != 0 && q->flags != taprio_flags) {
NL_SET_ERR_MSG_MOD(extack, "Changing 'flags' of a running schedule is not supported");
return -EOPNOTSUPP;
} else if (!FLAGS_VALID(taprio_flags)) {
NL_SET_ERR_MSG_MOD(extack, "Specified 'flags' are not valid");
return -EINVAL;
}
q->flags = taprio_flags;
}
err = taprio_parse_mqprio_opt(dev, mqprio, extack, taprio_flags);
if (err < 0) if (err < 0)
return err; return err;
...@@ -786,7 +1106,18 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt, ...@@ -786,7 +1106,18 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
/* Protects against enqueue()/dequeue() */ /* Protects against enqueue()/dequeue() */
spin_lock_bh(qdisc_lock(sch)); spin_lock_bh(qdisc_lock(sch));
if (!hrtimer_active(&q->advance_timer)) { if (tb[TCA_TAPRIO_ATTR_TXTIME_DELAY]) {
if (!TXTIME_ASSIST_IS_ENABLED(q->flags)) {
NL_SET_ERR_MSG_MOD(extack, "txtime-delay can only be set when txtime-assist mode is enabled");
err = -EINVAL;
goto unlock;
}
q->txtime_delay = nla_get_s32(tb[TCA_TAPRIO_ATTR_TXTIME_DELAY]);
}
if (!TXTIME_ASSIST_IS_ENABLED(taprio_flags) &&
!hrtimer_active(&q->advance_timer)) {
hrtimer_init(&q->advance_timer, q->clockid, HRTIMER_MODE_ABS); hrtimer_init(&q->advance_timer, q->clockid, HRTIMER_MODE_ABS);
q->advance_timer.function = advance_sched; q->advance_timer.function = advance_sched;
} }
...@@ -806,16 +1137,16 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt, ...@@ -806,16 +1137,16 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
switch (q->clockid) { switch (q->clockid) {
case CLOCK_REALTIME: case CLOCK_REALTIME:
q->get_time = ktime_get_real; q->tk_offset = TK_OFFS_REAL;
break; break;
case CLOCK_MONOTONIC: case CLOCK_MONOTONIC:
q->get_time = ktime_get; q->tk_offset = TK_OFFS_MAX;
break; break;
case CLOCK_BOOTTIME: case CLOCK_BOOTTIME:
q->get_time = ktime_get_boottime; q->tk_offset = TK_OFFS_BOOT;
break; break;
case CLOCK_TAI: case CLOCK_TAI:
q->get_time = ktime_get_clocktai; q->tk_offset = TK_OFFS_TAI;
break; break;
default: default:
NL_SET_ERR_MSG(extack, "Invalid 'clockid'"); NL_SET_ERR_MSG(extack, "Invalid 'clockid'");
...@@ -829,20 +1160,35 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt, ...@@ -829,20 +1160,35 @@ static int taprio_change(struct Qdisc *sch, struct nlattr *opt,
goto unlock; goto unlock;
} }
setup_first_close_time(q, new_admin, start); if (TXTIME_ASSIST_IS_ENABLED(taprio_flags)) {
setup_txtime(q, new_admin, start);
/* Protects against advance_sched() */ if (!oper) {
spin_lock_irqsave(&q->current_entry_lock, flags); rcu_assign_pointer(q->oper_sched, new_admin);
err = 0;
new_admin = NULL;
goto unlock;
}
taprio_start_sched(sch, start, new_admin); rcu_assign_pointer(q->admin_sched, new_admin);
if (admin)
call_rcu(&admin->rcu, taprio_free_sched_cb);
} else {
setup_first_close_time(q, new_admin, start);
rcu_assign_pointer(q->admin_sched, new_admin); /* Protects against advance_sched() */
if (admin) spin_lock_irqsave(&q->current_entry_lock, flags);
call_rcu(&admin->rcu, taprio_free_sched_cb);
new_admin = NULL;
spin_unlock_irqrestore(&q->current_entry_lock, flags); taprio_start_sched(sch, start, new_admin);
rcu_assign_pointer(q->admin_sched, new_admin);
if (admin)
call_rcu(&admin->rcu, taprio_free_sched_cb);
spin_unlock_irqrestore(&q->current_entry_lock, flags);
}
new_admin = NULL;
err = 0; err = 0;
unlock: unlock:
...@@ -1080,6 +1426,13 @@ static int taprio_dump(struct Qdisc *sch, struct sk_buff *skb) ...@@ -1080,6 +1426,13 @@ static int taprio_dump(struct Qdisc *sch, struct sk_buff *skb)
if (nla_put_s32(skb, TCA_TAPRIO_ATTR_SCHED_CLOCKID, q->clockid)) if (nla_put_s32(skb, TCA_TAPRIO_ATTR_SCHED_CLOCKID, q->clockid))
goto options_error; goto options_error;
if (q->flags && nla_put_u32(skb, TCA_TAPRIO_ATTR_FLAGS, q->flags))
goto options_error;
if (q->txtime_delay &&
nla_put_s32(skb, TCA_TAPRIO_ATTR_TXTIME_DELAY, q->txtime_delay))
goto options_error;
if (oper && dump_schedule(skb, oper)) if (oper && dump_schedule(skb, oper))
goto options_error; goto options_error;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment