public inbox for linux-staging@lists.linux.dev
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Kieran Bingham <kieran.bingham@ideasonboard.com>
Cc: Umang Jain <umang.jain@ideasonboard.com>,
	linux-staging@lists.linux.dev,
	Stefan Wahren <stefan.wahren@i2se.com>,
	Dan Carpenter <error27@gmail.com>,
	Phil Elwell <phil@raspberrypi.com>,
	Dave Stevenson <dave.stevenson@raspberrypi.com>
Subject: Re: [PATCH v2 3/6] staging: vc04_services: Drop global members for tracking connections
Date: Fri, 15 Mar 2024 13:50:09 +0200	[thread overview]
Message-ID: <20240315115009.GB3498@pendragon.ideasonboard.com> (raw)
In-Reply-To: <171042216625.252503.4862753149878092236@ping.linuxembedded.co.uk>

On Thu, Mar 14, 2024 at 01:16:06PM +0000, Kieran Bingham wrote:
> Quoting Umang Jain (2024-03-14 10:06:04)
> > Introduce a new struct vchiq_connected, responsible to track
> > the connections to the vchiq platform driver. The struct is added
> > as part of vchiq platform driver data (struct vchiq_drvdata).
> > 
> > Drop global members for tracking vchiq driver connections in
> > vchiq_connected.[ch] and use the struct vchiq_connected present
> > in struct vchiq_drvdata.
> > 
> > Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
> > ---
> >  .../interface/vchiq_arm/vchiq_arm.c           |  2 +-
> >  .../interface/vchiq_arm/vchiq_arm.h           |  3 ++
> >  .../interface/vchiq_arm/vchiq_connected.c     | 33 +++++++++----------
> >  .../interface/vchiq_arm/vchiq_connected.h     | 15 +++++++--
> >  4 files changed, 33 insertions(+), 20 deletions(-)
> > 
> > diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c
> > index 52569517ba4e..8c7520dee5f4 100644
> > --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c
> > +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.c
> > @@ -550,7 +550,7 @@ static int vchiq_platform_init(struct platform_device *pdev, struct vchiq_state
> >         dev_dbg(&pdev->dev, "arm: vchiq_init - done (slots %pK, phys %pad)\n",
> >                 vchiq_slot_zero, &slot_phys);
> >  
> > -       vchiq_call_connected_callbacks();
> > +       vchiq_call_connected_callbacks(&drvdata->drv_connected);
> >  
> >         return 0;
> >  }
> > diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.h b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.h
> > index 7fef05e7389c..f2fd572df2b3 100644
> > --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.h
> > +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_arm.h
> > @@ -16,6 +16,7 @@
> >  
> >  #include "vchiq_core.h"
> >  #include "vchiq_debugfs.h"
> > +#include "vchiq_connected.h"
> >  
> >  /* Some per-instance constants */
> >  #define MAX_COMPLETIONS 128
> > @@ -31,6 +32,8 @@ enum USE_TYPE_E {
> >  struct vchiq_drvdata {
> >         const unsigned int cache_line_size;
> >         struct rpi_firmware *fw;
> > +
> > +       struct vchiq_connected drv_connected;
> >  };
> >  
> >  struct user_service {
> > diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.c b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.c
> > index 4604a2f4d2de..bc21910fe823 100644
> > --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.c
> > +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.c
> > @@ -6,11 +6,6 @@
> >  #include <linux/module.h>
> >  #include <linux/mutex.h>
> >  
> > -#define  MAX_CALLBACKS  10
> > -
> > -static   int                        g_connected;
> > -static   int                        g_num_deferred_callbacks;
> > -static   void (*g_deferred_callback[MAX_CALLBACKS])(void);
> >  static   DEFINE_MUTEX(g_connected_mutex);
> >  
> >  /*
> > @@ -19,23 +14,27 @@ static   DEFINE_MUTEX(g_connected_mutex);
> >   * be made immediately, otherwise it will be deferred until
> >   * vchiq_call_connected_callbacks is called.
> >   */
> > -void vchiq_add_connected_callback(struct vchiq_device *device, void (*callback)(void))
> > +void vchiq_add_connected_callback(struct vchiq_device *device,
> > +                                 void (*callback)(void),
> > +                                 struct vchiq_connected *drv_connected)
> 
> I think I would have called this 'struct vchiq_connection *connection'
> but it doesn't really matter, so don't rework unless there's a 'need'
> for another iteration.

Yes, "connected" sounds weird.

This being said, I don't think we should add a new structure, and I also
don't think we should move this to vchiq_drvdata as-is.

vchiq_drvdata combines two things (in just two fields, quite an
achievement :-)):

- The cache_line_size field is static information about the platform. It
  should be moved to a vchiq_platform_info or similar structure, as
  that's what the .data field of the of_device_it entries should point
  to.

- The fw field is runtime data belonging to the vchiq "controller" or
  "bus" device (not sure how to name it). It's the equivalent, in a
  camera sensor driver for instance, of the private structure allocated
  per sensor instance at probe time. The vchiq_drvdata structure should
  be renamed accordingly (e.g. vchiq_controller, ...), and allocated
  dynamically in vchiq_probe().

The global variables related to connections can then move to that
controller structure, there's no need to create a sub-structure. It
should also not be passed explicitly to vchiq_add_connected_callback(),
but should be retrieved from vchiq_device. The vchiq_device structure
should store a pointer to vchiq_controller, which should be initialized
in vchiq_device_register(), using platform_get_drvdata(parent).

Does this make sense ?

> Aha though I see vchiq_connected matches the name of this c module, so
> maybe that's just as/more suitable.
> 
> 
> Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
> 
> >  {
> > +       unsigned int index;
> > +
> >         if (mutex_lock_killable(&g_connected_mutex))
> >                 return;
> >  
> > -       if (g_connected) {
> > +       if (drv_connected->connected) {
> >                 /* We're already connected. Call the callback immediately. */
> >                 callback();
> >         } else {
> > -               if (g_num_deferred_callbacks >= MAX_CALLBACKS) {
> > +               if (drv_connected->num_deferred_callbacks >= VCHIQ_DRV_MAX_CALLBACKS) {
> >                         dev_err(&device->dev,
> >                                 "core: There already %d callback registered - please increase MAX_CALLBACKS\n",
> > -                               g_num_deferred_callbacks);
> > +                               drv_connected->num_deferred_callbacks);
> >                 } else {
> > -                       g_deferred_callback[g_num_deferred_callbacks] =
> > -                               callback;
> > -                       g_num_deferred_callbacks++;
> > +                       index = drv_connected->num_deferred_callbacks;
> > +                       drv_connected->deferred_callback[index] = callback;
> > +                       drv_connected->num_deferred_callbacks++;
> >                 }
> >         }
> >         mutex_unlock(&g_connected_mutex);
> > @@ -46,17 +45,17 @@ EXPORT_SYMBOL(vchiq_add_connected_callback);
> >   * This function is called by the vchiq stack once it has been connected to
> >   * the videocore and clients can start to use the stack.
> >   */
> > -void vchiq_call_connected_callbacks(void)
> > +void vchiq_call_connected_callbacks(struct vchiq_connected *drv_connected)
> >  {
> >         int i;
> >  
> >         if (mutex_lock_killable(&g_connected_mutex))
> >                 return;
> >  
> > -       for (i = 0; i <  g_num_deferred_callbacks; i++)
> > -               g_deferred_callback[i]();
> > +       for (i = 0; i < drv_connected->num_deferred_callbacks; i++)
> > +               drv_connected->deferred_callback[i]();
> >  
> > -       g_num_deferred_callbacks = 0;
> > -       g_connected = 1;
> > +       drv_connected->num_deferred_callbacks = 0;
> > +       drv_connected->connected = true;
> >         mutex_unlock(&g_connected_mutex);
> >  }
> > diff --git a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.h b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.h
> > index e4ed56446f8a..66e56b5af46e 100644
> > --- a/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.h
> > +++ b/drivers/staging/vc04_services/interface/vchiq_arm/vchiq_connected.h
> > @@ -6,7 +6,18 @@
> >  #ifndef VCHIQ_CONNECTED_H
> >  #define VCHIQ_CONNECTED_H
> >  
> > -void vchiq_add_connected_callback(struct vchiq_device *device, void (*callback)(void));
> > -void vchiq_call_connected_callbacks(void);
> > +#define  VCHIQ_DRV_MAX_CALLBACKS  10
> > +
> > +struct vchiq_connected {
> > +       bool connected;
> > +       int num_deferred_callbacks;
> > +
> > +       void (*deferred_callback[VCHIQ_DRV_MAX_CALLBACKS])(void);
> > +};
> > +
> > +void vchiq_add_connected_callback(struct vchiq_device *device, void (*callback)(void),
> > +                                 struct vchiq_connected *drv_connected);
> > +void vchiq_call_connected_callbacks(struct vchiq_connected *drv_connected);
> > +
> >  
> >  #endif /* VCHIQ_CONNECTED_H */

-- 
Regards,

Laurent Pinchart

  reply	other threads:[~2024-03-15 11:50 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-14 10:06 [PATCH v2 0/6] staging: vc04_services: Drop non-essential global members Umang Jain
2024-03-14 10:06 ` [PATCH v2 1/6] staging: vc04_services: Drop g_once_init global variable Umang Jain
2024-03-14 13:00   ` Kieran Bingham
2024-03-15 10:58   ` Laurent Pinchart
2024-03-14 10:06 ` [PATCH v2 2/6] staging: vc04_services: vchiq_arm: Move struct vchiq_drvdata to header Umang Jain
2024-03-14 13:05   ` Kieran Bingham
2024-03-14 10:06 ` [PATCH v2 3/6] staging: vc04_services: Drop global members for tracking connections Umang Jain
2024-03-14 13:16   ` Kieran Bingham
2024-03-15 11:50     ` Laurent Pinchart [this message]
2024-03-14 10:06 ` [PATCH v2 4/6] staging: vc04_services: vchiq_arm: Drop g_cache_line_size Umang Jain
2024-03-14 14:51   ` Kieran Bingham
2024-03-14 15:39     ` Dan Carpenter
2024-03-14 10:06 ` [PATCH v2 5/6] staging: vc04_services: Drop global variables tracking allocated pages Umang Jain
2024-03-14 14:54   ` Kieran Bingham
2024-03-14 15:45     ` Dan Carpenter
2024-03-15  5:47     ` Umang Jain
2024-03-16  8:41       ` Dan Carpenter
2024-03-16 10:13         ` Kieran Bingham
2024-03-14 10:06 ` [PATCH v2 6/6] staging: vc04_services: Drop completed TODO item Umang Jain
2024-03-14 11:19 ` [PATCH v2 0/6] staging: vc04_services: Drop non-essential global members Dan Carpenter

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=20240315115009.GB3498@pendragon.ideasonboard.com \
    --to=laurent.pinchart@ideasonboard.com \
    --cc=dave.stevenson@raspberrypi.com \
    --cc=error27@gmail.com \
    --cc=kieran.bingham@ideasonboard.com \
    --cc=linux-staging@lists.linux.dev \
    --cc=phil@raspberrypi.com \
    --cc=stefan.wahren@i2se.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