lkml.org 
[lkml]   [2020]   [Oct]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH net-next 04/16] devlink: Add reload stats
From
Date

On 10/3/2020 12:00 PM, Jiri Pirko wrote:
> Thu, Oct 01, 2020 at 03:59:07PM CEST, moshe@mellanox.com wrote:
>> Add reload stats to hold the history per reload action type and limit.
>>
>> For example, the number of times fw_activate has been performed on this
>> device since the driver module was added or if the firmware activation
>> was performed with or without reset.
>>
>> Add devlink notification on stats update.
>>
>> Expose devlink reload stats to the user through devlink dev get command.
>>
>> Examples:
>> $ devlink dev show
>> pci/0000:82:00.0:
>> stats:
>> reload_stats:
>> driver_reinit 2
>> fw_activate 1
>> fw_activate_no_reset 0
>> pci/0000:82:00.1:
>> stats:
>> reload_stats:
>> driver_reinit 1
>> fw_activate 0
>> fw_activate_no_reset 0
>>
>> $ devlink dev show -jp
>> {
>> "dev": {
>> "pci/0000:82:00.0": {
>> "stats": {
>> "reload_stats": [ {
> Just "reload". No need to repeat "stats" here.
Ack.
>
>> "driver_reinit": 2
>> },{
>> "fw_activate": 1
>> },{
>> "fw_activate_no_reset": 0
>> } ]
>> }
>> },
>> "pci/0000:82:00.1": {
>> "stats": {
>> "reload_stats": [ {
>> "driver_reinit": 1
>> },{
>> "fw_activate": 0
>> },{
>> "fw_activate_no_reset": 0
>> } ]
>> }
>> }
>> }
>> }
>>
>> Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
>> ---
>> RFCv5 -> v1:
>> - Changed the stats output structure, have 2 stats, one for local and
>> one for remote
>> - Resplit this patch and the next one by remote/local reload stast
>> instead of set/get reload stats
>> - Add helper function devlink_reload_stats_put()
>> RFCv4 -> RFCv5:
>> - Add separate reload action stats for updating on remote actions
>> - Protect from updating remote actions stats during reload_down()/up()
>> RFCv3 -> RFCv4:
>> - Renamed reload_actions_cnts to reload_action_stats
>> - Add devlink notifications on stats update
>> - Renamed devlink_reload_actions_implicit_actions_performed() and add
>> function comment in code
>> RFCv2 -> RFCv3:
>> - New patch
>> ---
>> include/net/devlink.h | 7 +++
>> include/uapi/linux/devlink.h | 5 ++
>> net/core/devlink.c | 97 ++++++++++++++++++++++++++++++++++++
>> 3 files changed, 109 insertions(+)
>>
>> diff --git a/include/net/devlink.h b/include/net/devlink.h
>> index 43dde69086e5..0f3bd23b6c04 100644
>> --- a/include/net/devlink.h
>> +++ b/include/net/devlink.h
>> @@ -20,6 +20,9 @@
>> #include <uapi/linux/devlink.h>
>> #include <linux/xarray.h>
>>
>> +#define DEVLINK_RELOAD_STATS_ARRAY_SIZE \
>> + (__DEVLINK_RELOAD_LIMIT_MAX * __DEVLINK_RELOAD_ACTION_MAX)
>> +
>> struct devlink_ops;
>>
>> struct devlink {
>> @@ -38,6 +41,7 @@ struct devlink {
>> struct list_head trap_policer_list;
>> const struct devlink_ops *ops;
>> struct xarray snapshot_ids;
>> + u32 reload_stats[DEVLINK_RELOAD_STATS_ARRAY_SIZE];
>> struct device *dev;
>> possible_net_t _net;
>> struct mutex lock; /* Serializes access to devlink instance specific objects such as
>> @@ -1470,6 +1474,9 @@ void
>> devlink_health_reporter_recovery_done(struct devlink_health_reporter *reporter);
>>
>> bool devlink_is_reload_failed(const struct devlink *devlink);
>> +void devlink_remote_reload_actions_performed(struct devlink *devlink,
>> + enum devlink_reload_limit limit,
>> + unsigned long actions_performed);
> Leftover, please remove/move.
>

Yes, should be in the next patch, I missed it while re-splitting these
two patches.

>> void devlink_flash_update_begin_notify(struct devlink *devlink);
>> void devlink_flash_update_end_notify(struct devlink *devlink);
>> diff --git a/include/uapi/linux/devlink.h b/include/uapi/linux/devlink.h
>> index cc5dc4c07b4a..97e0137f6201 100644
>> --- a/include/uapi/linux/devlink.h
>> +++ b/include/uapi/linux/devlink.h
>> @@ -526,6 +526,11 @@ enum devlink_attr {
>> DEVLINK_ATTR_RELOAD_ACTIONS_PERFORMED, /* u64 */
>> DEVLINK_ATTR_RELOAD_LIMIT, /* u8 */
>>
>> + DEVLINK_ATTR_DEV_STATS, /* nested */
>> + DEVLINK_ATTR_RELOAD_STATS, /* nested */
>> + DEVLINK_ATTR_RELOAD_STATS_ENTRY, /* nested */
>> + DEVLINK_ATTR_RELOAD_STATS_VALUE, /* u32 */
>> +
>> /* add new attributes above here, update the policy in devlink.c */
>>
>> __DEVLINK_ATTR_MAX,
>> diff --git a/net/core/devlink.c b/net/core/devlink.c
>> index 6de7d6aa6ed1..05516f1e4c3e 100644
>> --- a/net/core/devlink.c
>> +++ b/net/core/devlink.c
>> @@ -500,10 +500,68 @@ devlink_reload_limit_is_supported(struct devlink *devlink, enum devlink_reload_l
>> return test_bit(limit, &devlink->ops->reload_limits);
>> }
>>
>> +static int devlink_reload_stat_put(struct sk_buff *msg, enum devlink_reload_action action,
>> + enum devlink_reload_limit limit, u32 value)
>> +{
>> + struct nlattr *reload_stats_entry;
>> +
>> + reload_stats_entry = nla_nest_start(msg, DEVLINK_ATTR_RELOAD_STATS_ENTRY);
>> + if (!reload_stats_entry)
>> + return -EMSGSIZE;
>> +
>> + if (nla_put_u8(msg, DEVLINK_ATTR_RELOAD_ACTION, action))
>> + goto nla_put_failure;
>> + if (nla_put_u8(msg, DEVLINK_ATTR_RELOAD_LIMIT, limit))
>> + goto nla_put_failure;
>> + if (nla_put_u32(msg, DEVLINK_ATTR_RELOAD_STATS_VALUE, value))
>> + goto nla_put_failure;
>> + nla_nest_end(msg, reload_stats_entry);
>> + return 0;
>> +
>> +nla_put_failure:
>> + nla_nest_cancel(msg, reload_stats_entry);
>> + return -EMSGSIZE;
>> +}
>> +
>> +static int devlink_reload_stats_put(struct sk_buff *msg, struct devlink *devlink)
>> +{
>> + struct nlattr *reload_stats_attr;
>> + int i, j, stat_idx;
>> + u32 value;
>> +
>> + reload_stats_attr = nla_nest_start(msg, DEVLINK_ATTR_RELOAD_STATS);
>> +
>> + if (!reload_stats_attr)
>> + return -EMSGSIZE;
>> +
>> + for (j = 0; j <= DEVLINK_RELOAD_LIMIT_MAX; j++) {
>> + if (j != DEVLINK_RELOAD_LIMIT_UNSPEC &&
> You should check limit_unspec during driver register, not here.


The thing is that now (change from RFCv5 to v1, see patch 3 change log)
driver does not need to register supported limits if it only supports
actions without any limitation. So I want to show stats of actions with
unspecified limit, though I don't expect driver to register it.

>
>> + !devlink_reload_limit_is_supported(devlink, j))
>> + continue;
>> + for (i = 0; i <= DEVLINK_RELOAD_ACTION_MAX; i++) {
>> + if (!devlink_reload_action_is_supported(devlink, i) ||
>> + devlink_reload_combination_is_invalid(i, j))
>> + continue;
>> +
>> + stat_idx = j * __DEVLINK_RELOAD_ACTION_MAX + i;
>> + value = devlink->reload_stats[stat_idx];
>> + if (devlink_reload_stat_put(msg, i, j, value))
>> + goto nla_put_failure;
>> + }
>> + }
>> + nla_nest_end(msg, reload_stats_attr);
>> + return 0;
>> +
>> +nla_put_failure:
>> + nla_nest_cancel(msg, reload_stats_attr);
>> + return -EMSGSIZE;
>> +}
>> +
>> static int devlink_nl_fill(struct sk_buff *msg, struct devlink *devlink,
>> enum devlink_command cmd, u32 portid,
>> u32 seq, int flags)
>> {
>> + struct nlattr *dev_stats;
>> void *hdr;
>>
>> hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
>> @@ -515,9 +573,19 @@ static int devlink_nl_fill(struct sk_buff *msg, struct devlink *devlink,
>> if (nla_put_u8(msg, DEVLINK_ATTR_RELOAD_FAILED, devlink->reload_failed))
>> goto nla_put_failure;
>>
>> + dev_stats = nla_nest_start(msg, DEVLINK_ATTR_DEV_STATS);
> Avoid the "DEV". Just "DEVLINK_ATTR_STATS" is enough.


"DEVLINK_ATTR_STATS" is already used for something else (traps), I added
DEV as its dev get command stats.

>
>> + if (!dev_stats)
>> + goto nla_put_failure;
>> +
>> + if (devlink_reload_stats_put(msg, devlink))
>> + goto dev_stats_nest_cancel;
>> +
>> + nla_nest_end(msg, dev_stats);
>> genlmsg_end(msg, hdr);
>> return 0;
>>
>> +dev_stats_nest_cancel:
>> + nla_nest_cancel(msg, dev_stats);
>> nla_put_failure:
>> genlmsg_cancel(msg, hdr);
>> return -EMSGSIZE;
>> @@ -3004,6 +3072,34 @@ bool devlink_is_reload_failed(const struct devlink *devlink)
>> }
>> EXPORT_SYMBOL_GPL(devlink_is_reload_failed);
>>
>> +static void
>> +__devlink_reload_stats_update(struct devlink *devlink, u32 *reload_stats,
>> + enum devlink_reload_limit limit, unsigned long actions_performed)
>> +{
>> + int stat_idx;
>> + int action;
>> +
>> + if (!actions_performed)
>> + return;
>> +
>> + if (WARN_ON(limit > DEVLINK_RELOAD_LIMIT_MAX))
> I don't understand the reason for this check and warn on. You should
> sanitize this in the caller (I think you already do that).


That's because the next patch has a call to this function directly from
the driver. I should add this check on the caller in the next patch.

>> + return;
>> + for (action = 0; action <= DEVLINK_RELOAD_ACTION_MAX; action++) {
>> + if (!test_bit(action, &actions_performed))
>> + continue;
>> + stat_idx = limit * __DEVLINK_RELOAD_ACTION_MAX + action;
>> + reload_stats[stat_idx]++;
>> + }
>> + devlink_notify(devlink, DEVLINK_CMD_NEW);
>> +}
>> +
>> +static void
>> +devlink_reload_stats_update(struct devlink *devlink, enum devlink_reload_limit limit,
>> + unsigned long actions_performed)
>> +{
>> + __devlink_reload_stats_update(devlink, devlink->reload_stats, limit, actions_performed);
>> +}
>> +
>> static int devlink_reload(struct devlink *devlink, struct net *dest_net,
>> enum devlink_reload_action action, enum devlink_reload_limit limit,
>> struct netlink_ext_ack *extack, unsigned long *actions_performed)
>> @@ -3026,6 +3122,7 @@ static int devlink_reload(struct devlink *devlink, struct net *dest_net,
>> return err;
>>
>> WARN_ON(!test_bit(action, actions_performed));
>> + devlink_reload_stats_update(devlink, limit, *actions_performed);
>> return 0;
>> }
>>
>> --
>> 2.18.2
>>

\
 
 \ /
  Last update: 2020-10-04 09:01    [W:0.088 / U:0.152 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site