Linux wireless drivers development
 help / color / mirror / Atom feed
From: Adham Abozaeid <adham.abozaeid@microchip.com>
To: Claudiu Beznea <Claudiu.Beznea@microchip.com>
Cc: Ajay Singh <ajay.kathat@microchip.com>,
	<linux-wireless@vger.kernel.org>, <devel@driverdev.osuosl.org>,
	<gregkh@linuxfoundation.org>, <ganesh.krishna@microchip.com>,
	<venkateswara.kaja@microchip.com>, <aditya.shankar@microchip.com>
Subject: Re: [PATCH 09/24] staging: wilc1000: move static variable clients_count to 'wilc' structure
Date: Fri, 24 Aug 2018 17:13:14 -0700	[thread overview]
Message-ID: <20180824171314.466fd48d@edcwlt13-vm> (raw)
In-Reply-To: <e56a014d-3df2-9977-8b5a-9793b2969ceb@microchip.com>

On Thu, 23 Aug 2018 11:09:56 +0300
Claudiu Beznea <Claudiu.Beznea@microchip.com> wrote:

> 
> 
> On 14.08.2018 09:50, Ajay Singh wrote:
> > Avoid use of static variable 'clients_count' and move it part of 'wilc'
> > structure.
> > 
> > Signed-off-by: Ajay Singh <ajay.kathat@microchip.com>
> > ---
> >  drivers/staging/wilc1000/host_interface.c     | 9 ++++-----
> >  drivers/staging/wilc1000/wilc_wfi_netdevice.h | 1 +
> >  2 files changed, 5 insertions(+), 5 deletions(-)
> > 
> > diff --git a/drivers/staging/wilc1000/host_interface.c b/drivers/staging/wilc1000/host_interface.c
> > index 6225e67..d930f06 100644
> > --- a/drivers/staging/wilc1000/host_interface.c
> > +++ b/drivers/staging/wilc1000/host_interface.c
> > @@ -199,7 +199,6 @@ static u8 rcv_assoc_resp[MAX_ASSOC_RESP_FRAME_SIZE];
> >  
> >  static u8 set_ip[2][4];
> >  static u8 get_ip[2][4];
> > -static u32 clients_count;
> >  
> >  static int host_int_get_ipaddress(struct wilc_vif *vif, u8 *ip_addr, u8 idx);
> >  
> > @@ -3456,7 +3455,7 @@ int wilc_init(struct net_device *dev, struct host_if_drv **hif_drv_handler)
> >  
> >  	vif->obtaining_ip = false;
> >  
> > -	if (clients_count == 0) {
> > +	if (wilc->clients_count == 0) {
> 
> The check of this should be mutex protected or a hif init/deinit should be
> done in probe/remove code not not in open (I'm for the second approach).
> Imagine this scenario:
> 
> vif 1: is down
> vif 2: is down. "ifconfig vif2 up" is executed and the execution reach this
>        point and then suspends after "if (wilc->clients_count == 0)" check
> vif 1: "ifconfig vif1 up" and execution reach the end of this function and
>        then the execution for vif 2 resumes. This will init twice the hif
>        mutex, completion etc.
> 
> >  		init_completion(&hif_driver_comp);
> >  		mutex_init(&hif_deinit_lock);
> >  

Instead of using mutex protection, it would be better to move this block to
wilc_netdev_init() where wilc structure is initialized 

> > @@ -3490,7 +3489,7 @@ int wilc_init(struct net_device *dev, struct host_if_drv **hif_drv_handler)
> >  
> >  	mutex_unlock(&hif_drv->cfg_values_lock);
> >  
> > -	clients_count++;
> > +	wilc->clients_count++;
> >  
> >  	return 0;
> >  }
> > @@ -3526,7 +3525,7 @@ int wilc_deinit(struct wilc_vif *vif)
> >  
> >  	hif_drv->hif_state = HOST_IF_IDLE;
> >  
> > -	if (clients_count == 1)	{
> > +	if (vif->wilc->clients_count == 1) {
> >  		struct host_if_msg *msg;
> >  
> >  		msg = wilc_alloc_work(vif, handle_hif_exit_work, true);
> > @@ -3544,7 +3543,7 @@ int wilc_deinit(struct wilc_vif *vif)
> >  
> >  	kfree(hif_drv);
> >  
> > -	clients_count--;
> > +	vif->wilc->clients_count--;
> >  	terminated_handle = NULL;
> >  	mutex_unlock(&hif_deinit_lock);
> >  	return result;
> > diff --git a/drivers/staging/wilc1000/wilc_wfi_netdevice.h b/drivers/staging/wilc1000/wilc_wfi_netdevice.h
> > index 8e56a28..8cccbbc 100644
> > --- a/drivers/staging/wilc1000/wilc_wfi_netdevice.h
> > +++ b/drivers/staging/wilc1000/wilc_wfi_netdevice.h
> > @@ -171,6 +171,7 @@ struct wilc {
> >  
> >  	struct rf_info dummy_statistics;
> >  	bool enable_ps;
> > +	int clients_count;
> >  };
> >  
> >  struct wilc_wfi_mon_priv {
> > 

  reply	other threads:[~2018-08-25  3:50 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-14  6:49 [PATCH 00/24] staging: wilc1000: avoid use of static and global variable Ajay Singh
2018-08-14  6:49 ` [PATCH 01/24] staging: wilc1000: move 'wilc_enable_ps' global variable into 'wilc' struct Ajay Singh
2018-08-14  6:49 ` [PATCH 02/24] staging: wilc1000: move 'aging_timer' static variable to wilc_priv struct Ajay Singh
2018-08-14  6:49 ` [PATCH 03/24] staging: wilc1000: fix to use correct index to free scanned info in clear_shadow_scan() Ajay Singh
2018-08-14  6:49 ` [PATCH 04/24] staging: wilc1000: remove unnecessary NULL check " Ajay Singh
2018-08-14  6:49 ` [PATCH 05/24] staging: wilc1000: moved last_scanned_shadow & last_scanned_cnt to wilc_priv struct Ajay Singh
2018-08-14  6:49 ` [PATCH 06/24] staging: wilc1000: move during_ip_timer & wilc_optaining_ip to 'wilc_vif' struct Ajay Singh
2018-08-23  8:09   ` Claudiu Beznea
2018-08-23  9:43     ` Ajay Singh
2018-08-24  8:47       ` Claudiu Beznea
2018-08-14  6:49 ` [PATCH 07/24] staging: wilc1000: remove unused variable 'op_ifcs' Ajay Singh
2018-08-14  6:50 ` [PATCH 08/24] staging: wilc1000: avoid use of extra 'if' condition in wilc_init() Ajay Singh
2018-08-14  6:50 ` [PATCH 09/24] staging: wilc1000: move static variable clients_count to 'wilc' structure Ajay Singh
2018-08-23  8:09   ` Claudiu Beznea
2018-08-25  0:13     ` Adham Abozaeid [this message]
2018-08-14  6:50 ` [PATCH 10/24] staging: wilc1000: move wilc_multicast_mac_addr_list to 'wilc_vif' struct Ajay Singh
2018-08-23  8:10   ` Claudiu Beznea
2018-08-23 10:00     ` Ajay Singh
2018-08-24  8:47       ` Claudiu Beznea
2018-08-25  0:32         ` Adham Abozaeid
2018-08-27  5:40           ` Ajay Singh
2018-08-14  6:50 ` [PATCH 11/24] staging: wilc1000: move hif specific static variables to 'wilc' structure Ajay Singh
2018-08-23  8:11   ` Claudiu Beznea
2018-08-23 10:09     ` Ajay Singh
2018-08-24  8:47       ` Claudiu Beznea
2018-08-14  6:50 ` [PATCH 12/24] staging: wilc1000: move static variable 'terminated_handle' to wilc_vif struct Ajay Singh
2018-08-23  8:11   ` Claudiu Beznea
2018-08-23 14:36     ` Ajay Singh
2018-08-24  8:46       ` Claudiu Beznea
2018-08-27  5:27         ` Ajay Singh
2018-08-14  6:50 ` [PATCH 13/24] staging: wilc1000: move 'periodic_rssi' as part of 'wilc_vif' struct Ajay Singh
2018-08-14  6:50 ` [PATCH 14/24] staging: wilc1000: rename 'dummy_statistics' variable to 'periodic_stat' Ajay Singh
2018-08-14  6:50 ` [PATCH 15/24] staging: wilc1000: move 'rcv_assoc_resp' as part of hif_drv Ajay Singh
2018-08-14  6:50 ` [PATCH 16/24] staging: wilc1000: refactor tcp_process() to avoid extra leading tabs Ajay Singh
2018-08-14  6:50 ` [PATCH 17/24] staging: wilc1000: use lowercase for get_BSSID() and HIL variable Ajay Singh
2018-08-14  6:50 ` [PATCH 18/24] staging: wilc1000: move tcp_ack_filter algo related variables to 'wilc_vif' struct Ajay Singh
2018-08-14  6:50 ` [PATCH 19/24] staging: wilc1000: avoid line over 80 chars in wilc_wlan_txq_filter_dup_tcp_ack() Ajay Singh
2018-08-23  8:11   ` Claudiu Beznea
2018-08-23 12:18     ` Ajay Singh
2018-08-14  6:50 ` [PATCH 20/24] staging: wilc1000: avoid line over 80 chars in tcp_process() Ajay Singh
2018-08-23  8:12   ` Claudiu Beznea
2018-08-23 10:33     ` Ajay Singh
2018-08-24  9:31       ` Claudiu Beznea
2018-08-27  5:24         ` Ajay Singh
2018-08-27 12:00           ` Dan Carpenter
2018-08-28  4:29             ` Ajay Singh
2018-08-14  6:50 ` [PATCH 21/24] staging: wilc1000: remove unused code to set and get IP address Ajay Singh
2018-08-14  6:50 ` [PATCH 22/24] staging: wilc1000: move 'chip_ps_state' static variable as part of 'wilc' struct Ajay Singh
2018-08-14  6:50 ` [PATCH 23/24] staging: wilc1000: move 'wilc_connecting' static variable to 'wilc_vif' struct Ajay Singh
2018-08-23  8:12   ` Claudiu Beznea
2018-08-23 10:55   ` Greg KH
2018-08-23 11:27     ` Ajay Singh
2018-08-23 12:37       ` Dan Carpenter
2018-08-23 13:06         ` Ajay Singh
2018-08-14  6:50 ` [PATCH 24/24] staging: wilc1000: remove unnecessary static variable 'p2p_listen_state' Ajay Singh
2018-08-23  8:13   ` Claudiu Beznea
2018-08-23 12:07     ` Ajay Singh
2018-08-23  8:09 ` [PATCH 00/24] staging: wilc1000: avoid use of static and global variable Claudiu Beznea
2018-08-23  9:35   ` Ajay Singh
2018-08-27 17:10 ` Greg KH
2018-08-28  4:35   ` Ajay Singh

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=20180824171314.466fd48d@edcwlt13-vm \
    --to=adham.abozaeid@microchip.com \
    --cc=Claudiu.Beznea@microchip.com \
    --cc=aditya.shankar@microchip.com \
    --cc=ajay.kathat@microchip.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=ganesh.krishna@microchip.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=venkateswara.kaja@microchip.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