From: Martin Wilck <martin.wilck@suse.com>
To: "bmarzins@redhat.com" <bmarzins@redhat.com>,
"christophe.varoqui@opensvc.com" <christophe.varoqui@opensvc.com>
Cc: "dm-devel@redhat.com" <dm-devel@redhat.com>
Subject: Re: [dm-devel] [PATCH 3/4] multipathd: pass in the type of reconfigure
Date: Mon, 15 Nov 2021 16:23:54 +0000 [thread overview]
Message-ID: <0b1c32c80fb5263df6263570f26b60247576f1b5.camel@suse.com> (raw)
In-Reply-To: <1632180076-9294-4-git-send-email-bmarzins@redhat.com>
On Mon, 2021-09-20 at 18:21 -0500, Benjamin Marzinski wrote:
> This code doesn't actually change how multipathd reconfigures. It still
> does a weak reconfigure at the start, and full reconfigures later.
AFAICS it does change this, at least for the cases where reconfigure()
is called from ev_add_map() and missing_uev_wait_tick, where you call
schedule_reconfigure(false). The old code was using a static variable
and thus always doing a hard reconfigure when called more than once.
I'm find with this change (the old behavior was arguably wrong), but
the commit message should be clarified in this respect.
> But
> now schedule_reconfigure() takes the type of reconfigure to do, and
> that
> gets passed down to reconfigure(). If a full reconfigure has already
> been requested but not started, weak reconfigure requests will be
> upgraded. A future patch will enable users to control what kind of
> reconfigures happen.
>
> Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
> ---
> libmultipath/configure.c | 2 +-
> multipathd/cli_handlers.c | 2 +-
> multipathd/main.c | 62 ++++++++++++++++++++-------------------
> multipathd/main.h | 2 +-
> 4 files changed, 35 insertions(+), 33 deletions(-)
>
> diff --git a/libmultipath/configure.c b/libmultipath/configure.c
> index 7edb355b..eb8ec1bd 100644
> --- a/libmultipath/configure.c
> +++ b/libmultipath/configure.c
> @@ -1098,7 +1098,7 @@ out:
> * FORCE_RELOAD_NONE: existing maps aren't touched at all
> * FORCE_RELOAD_YES: all maps are rebuilt from scratch and (re)loaded
> in DM
> * FORCE_RELOAD_WEAK: existing maps are compared to the current conf
> and only
> - * reloaded in DM if there's a difference. This is useful during
> startup.
> + * reloaded in DM if there's a difference. This is normally
> sufficient.
> */
> int coalesce_paths (struct vectors *vecs, vector mpvec, char *refwwid,
> int force_reload, enum mpath_cmds cmd)
> diff --git a/multipathd/cli_handlers.c b/multipathd/cli_handlers.c
> index f59db3ab..b12a4e7e 100644
> --- a/multipathd/cli_handlers.c
> +++ b/multipathd/cli_handlers.c
> @@ -1012,7 +1012,7 @@ cli_reconfigure(void * v, struct strbuf *reply,
> void * data)
> {
> condlog(2, "reconfigure (operator)");
>
> - schedule_reconfigure();
> + schedule_reconfigure(true);
> return 0;
> }
>
> diff --git a/multipathd/main.c b/multipathd/main.c
> index 1ead0904..5c831e8d 100644
> --- a/multipathd/main.c
> +++ b/multipathd/main.c
> @@ -162,13 +162,6 @@ static bool get_delayed_reconfig(void)
> return val;
> }
>
> -static void set_delayed_reconfig(bool val)
> -{
> - pthread_mutex_lock(&config_lock);
> - __delayed_reconfig = val;
> - pthread_mutex_unlock(&config_lock);
> -}
> -
> /*
> * global copy of vecs for use in sig handlers
> */
> @@ -290,7 +283,18 @@ enum daemon_status wait_for_state_change_if(enum
> daemon_status oldstate,
> }
>
> /* Don't access this variable without holding config_lock */
> -static bool reconfigure_pending;
> +static enum force_reload_types reconfigure_pending =
> FORCE_RELOAD_NONE;
> +/* Only set while changing to DAEMON_CONFIGURE, and only access while
> + * reconfiguring in DAEMON_CONFIGURE */
> +static volatile enum force_reload_types reload_type =
> FORCE_RELOAD_NONE;
> +
> +static void enable_delayed_reconfig(enum force_reload_types type)
> +{
> + pthread_mutex_lock(&config_lock);
> + reconfigure_pending = type;
> + __delayed_reconfig = true;
> + pthread_mutex_unlock(&config_lock);
> +}
>
> /* must be called with config_lock held */
> static void __post_config_state(enum daemon_status state)
> @@ -305,7 +309,8 @@ static void __post_config_state(enum daemon_status
> state)
> * In either case, child() will see DAEMON_CONFIGURE
> * again and start another reconfigure cycle.
> */
> - if (reconfigure_pending && state == DAEMON_IDLE &&
> + if (reconfigure_pending != FORCE_RELOAD_NONE &&
> + state == DAEMON_IDLE &&
> (old_state == DAEMON_CONFIGURE ||
> old_state == DAEMON_RUNNING)) {
> /*
> @@ -317,7 +322,8 @@ static void __post_config_state(enum daemon_status
> state)
> state = DAEMON_CONFIGURE;
> }
> if (state == DAEMON_CONFIGURE) {
> - reconfigure_pending = false;
> + reload_type = (reconfigure_pending ==
> FORCE_RELOAD_YES) ? FORCE_RELOAD_YES : FORCE_RELOAD_WEAK;
> + reconfigure_pending = FORCE_RELOAD_NONE;
> __delayed_reconfig = false;
> }
> running_state = state;
> @@ -334,20 +340,25 @@ void post_config_state(enum daemon_status state)
> pthread_cleanup_pop(1);
> }
>
> -void schedule_reconfigure(void)
> +void schedule_reconfigure(bool force_reload_yes)
> {
> pthread_mutex_lock(&config_lock);
> pthread_cleanup_push(config_cleanup, NULL);
> + enum force_reload_types type;
> +
> + type = (reconfigure_pending == FORCE_RELOAD_YES ||
> force_reload_yes) ?
> + FORCE_RELOAD_YES : FORCE_RELOAD_WEAK;
> switch (running_state)
> {
> case DAEMON_SHUTDOWN:
> break;
> case DAEMON_IDLE:
> + reconfigure_pending = type;
> __post_config_state(DAEMON_CONFIGURE);
> break;
> case DAEMON_CONFIGURE:
> case DAEMON_RUNNING:
> - reconfigure_pending = true;
> + reconfigure_pending = type;
> break;
> default:
> break;
> @@ -812,7 +823,7 @@ ev_add_map (char * dev, const char * alias, struct
> vectors * vecs)
> if (get_delayed_reconfig() &&
> !need_to_delay_reconfig(vecs)) {
> condlog(2, "reconfigure (delayed)");
> - schedule_reconfigure();
> + schedule_reconfigure(false);
> return 0;
> }
> }
> @@ -1830,7 +1841,7 @@ missing_uev_wait_tick(struct vectors *vecs)
> if (timed_out && get_delayed_reconfig() &&
> !need_to_delay_reconfig(vecs)) {
> condlog(2, "reconfigure (delayed)");
> - schedule_reconfigure();
> + schedule_reconfigure(false);
> }
> }
>
> @@ -2588,14 +2599,13 @@ checkerloop (void *ap)
> }
>
> int
> -configure (struct vectors * vecs)
> +configure (struct vectors * vecs, enum force_reload_types type)
> {
> struct multipath * mpp;
> struct path * pp;
> vector mpvec;
> int i, ret;
> struct config *conf;
> - static int force_reload = FORCE_RELOAD_WEAK;
>
> if (!vecs->pathvec && !(vecs->pathvec = vector_alloc())) {
> condlog(0, "couldn't allocate path vec in configure");
> @@ -2643,15 +2653,7 @@ configure (struct vectors * vecs)
> if (should_exit())
> goto fail;
>
> - /*
> - * create new set of maps & push changed ones into dm
> - * In the first call, use FORCE_RELOAD_WEAK to avoid making
> - * superfluous ACT_RELOAD ioctls. Later calls are done
> - * with FORCE_RELOAD_YES.
> - */
> - ret = coalesce_paths(vecs, mpvec, NULL, force_reload,
> CMD_NONE);
> - if (force_reload == FORCE_RELOAD_WEAK)
> - force_reload = FORCE_RELOAD_YES;
> + ret = coalesce_paths(vecs, mpvec, NULL, type, CMD_NONE);
Can we please rename this variable from "type" to "reload_type" here,
for the sake of readability? Likewise, in reconfigure()?
> if (ret != CP_OK) {
> condlog(0, "configure failed while coalescing paths");
> goto fail;
> @@ -2729,7 +2731,7 @@ void rcu_free_config(struct rcu_head *head)
> }
>
> int
> -reconfigure (struct vectors * vecs)
> +reconfigure (struct vectors * vecs, enum force_reload_types type)
> {
> struct config * old, *conf;
>
> @@ -2763,7 +2765,7 @@ reconfigure (struct vectors * vecs)
> rcu_assign_pointer(multipath_conf, conf);
> call_rcu(&old->rcu, rcu_free_config);
>
> - configure(vecs);
> + configure(vecs, type);
>
>
> return 0;
> @@ -2815,7 +2817,7 @@ handle_signals(bool nonfatal)
> return;
> if (reconfig_sig) {
> condlog(2, "reconfigure (signal)");
> - schedule_reconfigure();
> + schedule_reconfigure(true);
> }
> if (log_reset_sig) {
> condlog(2, "reset log (signal)");
> @@ -3274,9 +3276,9 @@ child (__attribute__((unused)) void *param)
> lock(&vecs->lock);
> pthread_testcancel();
> if (!need_to_delay_reconfig(vecs))
> - reconfigure(vecs);
> + reconfigure(vecs, reload_type);
> else
> - set_delayed_reconfig(true);
> + enable_delayed_reconfig(reload_type);
> lock_cleanup_pop(vecs->lock);
> post_config_state(DAEMON_IDLE);
> }
> diff --git a/multipathd/main.h b/multipathd/main.h
> index a1697a74..c8a1ce92 100644
> --- a/multipathd/main.h
> +++ b/multipathd/main.h
> @@ -37,7 +37,7 @@ void exit_daemon(void);
> const char * daemon_status(void);
> enum daemon_status wait_for_state_change_if(enum daemon_status
> oldstate,
> unsigned long ms);
> -void schedule_reconfigure(void);
> +void schedule_reconfigure(bool reconfig_all);
For consistency, please consider using an "enum force_reload_types" as
argument to schedule_reconfigure() rather than a boolean.
Otherwise, the patch looks good.
Cheers
Martin
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
next prev parent reply other threads:[~2021-11-15 16:27 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-20 23:21 [dm-devel] [PATCH 0/4] Add "reconfigure all" multipath command Benjamin Marzinski
2021-09-20 23:21 ` [dm-devel] [PATCH 1/4] multipathd: move delayed_reconfig out of struct config Benjamin Marzinski
2021-11-15 16:10 ` Martin Wilck
2021-11-15 16:27 ` Martin Wilck
2021-11-15 23:45 ` Benjamin Marzinski
2021-09-20 23:21 ` [dm-devel] [PATCH 2/4] multipathd: remove reconfigure from header file Benjamin Marzinski
2021-11-15 16:10 ` Martin Wilck
2021-09-20 23:21 ` [dm-devel] [PATCH 3/4] multipathd: pass in the type of reconfigure Benjamin Marzinski
2021-11-15 16:23 ` Martin Wilck [this message]
2021-11-15 23:46 ` Benjamin Marzinski
2021-09-20 23:21 ` [dm-devel] [PATCH 4/4] multipathd: add "reconfigure all" command Benjamin Marzinski
2021-11-15 16:30 ` Martin Wilck
2021-09-24 20:44 ` [dm-devel] [PATCH 0/4] Add "reconfigure all" multipath command Xose Vazquez Perez
2021-09-27 15:11 ` Benjamin Marzinski
2021-09-28 8:25 ` Martin Wilck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=0b1c32c80fb5263df6263570f26b60247576f1b5.camel@suse.com \
--to=martin.wilck@suse.com \
--cc=bmarzins@redhat.com \
--cc=christophe.varoqui@opensvc.com \
--cc=dm-devel@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).