From: Stefan Wahren <wahrenst@gmx.net>
To: Umang Jain <umang.jain@ideasonboard.com>, linux-staging@lists.linux.dev
Cc: Dan Carpenter <error27@gmail.com>,
Kieran Bingham <kieran.bingham@ideasonboard.com>,
Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
Dave Stevenson <dave.stevenson@raspberrypi.com>,
Phil Elwell <phil@raspberrypi.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: [PATCH v2 1/2] staging: vchiq_core: Bubble up wait_event_interruptible() return value
Date: Wed, 3 Jul 2024 18:18:19 +0200 [thread overview]
Message-ID: <2a5c162e-0036-4f3e-b946-bce549e08c54@gmx.net> (raw)
In-Reply-To: <20240703131052.597443-2-umang.jain@ideasonboard.com>
Hi Umang,
Am 03.07.24 um 15:10 schrieb Umang Jain:
> wait_event_interruptible() returns if the condition evaluates to true
> it receives a signal. However, the current code always assume that the
> wait_event_interruptible() returns only when the event is fired.
> This should not be the case as wait_event_interruptible() can
> return on receiving a signal (with -ERESTARTSYS as return value).
>
> We should consider this and bubble up the return value of
> wait_event_interruptible() to exactly know if the wait has failed
> and error out. This will also help to properly stop kthreads in the
> subsequent patch.
>
> Meanwhile at it, remote_wait_event() is modified to return 0 on success,
> and an error code (from wait_event_interruptible()) on failure. The
> return value is now checked for remote_wait_event() calls.
>
> Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
> ---
> .../interface/vchiq_arm/vchiq_core.c | 31 ++++++++++++++-----
> 1 file changed, 24 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> index 4f65e4021c4d..dd70f2881eca 100644
> --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_core.c
> @@ -501,16 +501,21 @@ remote_event_create(wait_queue_head_t *wq, struct remote_event *event)
> * routines where switched to the "interruptible" family of functions, as the
> * former was deemed unjustified and the use "killable" set all VCHIQ's
> * threads in D state.
> + *
> + * Returns: 0 on success, a negative error code on failure
> */
> static inline int
> remote_event_wait(wait_queue_head_t *wq, struct remote_event *event)
> {
> + int ret = 0;
> +
> if (!event->fired) {
> event->armed = 1;
> dsb(sy);
> - if (wait_event_interruptible(*wq, event->fired)) {
> + ret = wait_event_interruptible(*wq, event->fired);
> + if (ret) {
> event->armed = 0;
> - return 0;
> + return ret;
> }
> event->armed = 0;
> /* Ensure that the peer sees that we are not waiting (armed == 0). */
> @@ -518,7 +523,7 @@ remote_event_wait(wait_queue_head_t *wq, struct remote_event *event)
> }
>
> event->fired = 0;
> - return 1;
> + return ret;
in general this patch looks good to me. But maybe we better return 0
directly and reduce the scope of ret.
> }
>
> /*
> @@ -1140,6 +1145,7 @@ queue_message_sync(struct vchiq_state *state, struct vchiq_service *service,
> struct vchiq_header *header;
> ssize_t callback_result;
> int svc_fourcc;
> + int ret;
>
> local = state->local;
>
> @@ -1147,7 +1153,9 @@ queue_message_sync(struct vchiq_state *state, struct vchiq_service *service,
> mutex_lock_killable(&state->sync_mutex))
> return -EAGAIN;
>
> - remote_event_wait(&state->sync_release_event, &local->sync_release);
> + ret = remote_event_wait(&state->sync_release_event, &local->sync_release);
> + if (ret)
> + return ret;
>
> /* Ensure that reads don't overtake the remote_event_wait. */
> rmb();
> @@ -1929,13 +1937,16 @@ slot_handler_func(void *v)
> {
> struct vchiq_state *state = v;
> struct vchiq_shared_state *local = state->local;
> + int ret;
>
> DEBUG_INITIALISE(local);
>
> while (1) {
> DEBUG_COUNT(SLOT_HANDLER_COUNT);
> DEBUG_TRACE(SLOT_HANDLER_LINE);
> - remote_event_wait(&state->trigger_event, &local->trigger);
> + ret = remote_event_wait(&state->trigger_event, &local->trigger);
> + if (ret)
> + return ret;
>
> /* Ensure that reads don't overtake the remote_event_wait. */
> rmb();
> @@ -1966,6 +1977,7 @@ recycle_func(void *v)
> struct vchiq_shared_state *local = state->local;
> u32 *found;
> size_t length;
> + int ret;
>
> length = sizeof(*found) * BITSET_SIZE(VCHIQ_MAX_SERVICES);
>
> @@ -1975,7 +1987,9 @@ recycle_func(void *v)
> return -ENOMEM;
>
> while (1) {
> - remote_event_wait(&state->recycle_event, &local->recycle);
> + ret = remote_event_wait(&state->recycle_event, &local->recycle);
> + if (ret)
> + return ret;
>
> process_free_queue(state, found, length);
> }
> @@ -1992,6 +2006,7 @@ sync_func(void *v)
> (struct vchiq_header *)SLOT_DATA_FROM_INDEX(state,
> state->remote->slot_sync);
> int svc_fourcc;
> + int ret;
>
> while (1) {
> struct vchiq_service *service;
> @@ -1999,7 +2014,9 @@ sync_func(void *v)
> int type;
> unsigned int localport, remoteport;
>
> - remote_event_wait(&state->sync_trigger_event, &local->sync_trigger);
> + ret = remote_event_wait(&state->sync_trigger_event, &local->sync_trigger);
> + if (ret)
> + return ret;
>
> /* Ensure that reads don't overtake the remote_event_wait. */
> rmb();
next prev parent reply other threads:[~2024-07-03 16:18 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-03 13:10 [PATCH v2 0/2] staging: vchiq_core: Stop kthreads on module unload Umang Jain
2024-07-03 13:10 ` [PATCH v2 1/2] staging: vchiq_core: Bubble up wait_event_interruptible() return value Umang Jain
2024-07-03 16:18 ` Stefan Wahren [this message]
2024-07-03 16:54 ` Dan Carpenter
2024-07-03 13:10 ` [PATCH v2 2/2] staging: vc04_services: vchiq_core: Stop kthreads on vchiq module unload Umang Jain
2024-07-03 16:31 ` Stefan Wahren
2024-07-03 16:44 ` Umang Jain
2024-07-06 8:50 ` [PATCH v2 0/2] staging: vchiq_core: Stop kthreads on " Stefan Wahren
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=2a5c162e-0036-4f3e-b946-bce549e08c54@gmx.net \
--to=wahrenst@gmx.net \
--cc=dave.stevenson@raspberrypi.com \
--cc=error27@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=kieran.bingham@ideasonboard.com \
--cc=laurent.pinchart@ideasonboard.com \
--cc=linux-staging@lists.linux.dev \
--cc=phil@raspberrypi.com \
--cc=umang.jain@ideasonboard.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