linux-kernel - Re: [PATCH] sound/core/seq: remove dead code

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <s5hv9emkxfc.wl-tiwai@suse.de>
Date:   Tue, 03 Nov 2020 21:17:11 +0100
From:   Takashi Iwai <tiwai@...e.de>
To:     Yu Hao <yuhaobehappy@...il.com>
Cc:     perex@...ex.cz, tiwai@...e.com, alsa-devel@...a-project.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sound/core/seq: remove dead code

On Tue, 03 Nov 2020 19:43:55 +0100,
Yu Hao wrote:
> 
> Yes, I can do that.
> 
> And a quick question, to resubmit with that change means that I should create
> a new submission and this one ends, right?

Yes, just submit a fresh patch.


thanks,

Takashi

> 
> Thanks.
> 
> On Tue, Nov 3, 2020 at 5:09 AM Takashi Iwai <tiwai@...e.de> wrote:
> 
>     On Mon, 02 Nov 2020 08:11:54 +0100,
>     Yu Hao wrote:
>     >
>     > The function snd_seq_queue_client_termination() is only called from
>     > function seq_free_client1(). The seq_free_client1() calls function
>     > snd_seq_queue_client_leave(), which deletes all objects whose owner
>     > equals to client->number in global array queue_list, before the function
>     > snd_seq_queue_client_termination(), which checks whether there are
>     > any objects in global array queue_list whose owner equals to
>     > client->number, with the same argument client->number. So
>     > the checking code in function snd_seq_queue_client_termination() is
>     > dead code. Remove those dead code.
>     >
>     > Signed-off-by: Yu Hao <yuhaobehappy@...il.com>
>    
>     Actually the whole function snd_seq_queue_client_termination() can be
>     removed.  It's a quite old code and I don't remember clearly, but I
>     guess the intention was to call this before the actual queue
>     deletion.  OTOH, it doesn't make any sense to change that order any
>     longer, as the snd_seq_queue_client_leave() already takes similar
>     procedure anyway, so let's rather wipe off the useless function.
>    
>     Could you resubmit with that change?
> 
>     thanks,
>    
>     Takashi
>    
>     > ---
>     >  sound/core/seq/seq_queue.c | 11 -----------
>     >  1 file changed, 11 deletions(-)
>     >
>     > diff --git a/sound/core/seq/seq_queue.c b/sound/core/seq/seq_queue.c
>     > index 71a6ea62c3be..91b3f3295d0b 100644
>     > --- a/sound/core/seq/seq_queue.c
>     > +++ b/sound/core/seq/seq_queue.c
>     > @@ -545,21 +545,10 @@ void snd_seq_queue_client_termination(int client)
>     >       unsigned long flags;
>     >       int i;
>     >       struct snd_seq_queue *q;
>     > -     bool matched;
>     > 
>     >       for (i = 0; i < SNDRV_SEQ_MAX_QUEUES; i++) {
>     >               if ((q = queueptr(i)) == NULL)
>     >                       continue;
>     > -             spin_lock_irqsave(&q->owner_lock, flags);
>     > -             matched = (q->owner == client);
>     > -             if (matched)
>     > -                     q->klocked = 1;
>     > -             spin_unlock_irqrestore(&q->owner_lock, flags);
>     > -             if (matched) {
>     > -                     if (q->timer->running)
>     > -                             snd_seq_timer_stop(q->timer);
>     > -                     snd_seq_timer_reset(q->timer);
>     > -             }
>     >               queuefree(q);
>     >       }
>     >  }
>     > --
>     > 2.17.1
>     >
> 
>