linux-kernel - Re: [alsa-devel] Question about of_clk_put ?

Open Source and information security mailing list archives

Message-ID: <87r31fee05.wl%kuninori.morimoto.gx@renesas.com>
Date:   Thu, 30 Mar 2017 01:46:42 +0000
From:   Kuninori Morimoto <kuninori.morimoto.gx@...esas.com>
To:     Kuninori Morimoto <kuninori.morimoto.gx@...esas.com>
CC:     Stephen Boyd <sboyd@...eaurora.org>,
        Russell King <linux@...linux.org.uk>,
        Linux-Renesas <linux-renesas-soc@...r.kernel.org>,
        Linux-ALSA <alsa-devel@...a-project.org>,
        Linux-Kernel <linux-kernel@...r.kernel.org>,
        <linux-clk@...r.kernel.org>
Subject: Re: [alsa-devel] Question about of_clk_put ?

Hi

Sorry for my noise.
I could solve my issue.

> Now, I'm using devm_get_clk_from_child() (= linux/drivers/clk/clk-devres.c)
> and I got Oops if I do bind/unbind driver several times.
> 
> 	[   32.008847] Unable to handle kernel paging request at virtual address d503201faa1e03e0
> 	[   32.017124] pgd = ffff8006f9060000
> 	[   32.020883] [d503201faa1e03e0] *pgd=0000000000000000
> 	[   32.026243] Internal error: Oops: 96000004 [#1] PREEMPT SMP
> 	[   32.032198] CPU: 0 PID: 934 Comm: kworker/0:2 Not tainted 4.11.0-rc3+ #1259
> 	[   32.039573] Hardware name: Renesas Salvator-X board based on r8a7795 (DT)
> 	[   32.046814] Workqueue: events deferred_probe_work_func
> 	[   32.052405] task: ffff8006fad2d800 task.stack: ffff8006f90b8000
> 	[   32.058809] PC is at __of_clk_get_from_provider+0x174/0x1b0
> 	[   32.064878] LR is at __of_clk_get_from_provider+0x164/0x1b0
> 	....
> 	[   32.746677] [<ffff0000083b581c>] __of_clk_get_from_provider+0x174/0x1b0
> 	[   32.754131] [<ffff0000083aded4>] __of_clk_get_by_name+0x104/0x140
> 	[   32.761058] [<ffff0000083adfd8>] of_clk_get_by_name+0x30/0x50
> 	[   32.767630] [<ffff0000083add1c>] devm_get_clk_from_child+0x54/0xb0
> 	...
> 
> I tried to find the criminal point, but, I couldn't specify where it is.
> Sometimes it is NULL pointer access crashed,
> sometimes it is crashed on of_clk_src_onecell_get(),
> sometimes there is no Oops.
> I want to solve this issue, but I want to know about of_clk_put as 1st step.
> 
> In devm_clk_get() case, it is using clk_get() <-> clk_put() pair, this is OK.
> In devm_get_clk_from_child() case, it is using of_clk_get_by_name() (= __of_clk_get())
> <-> clk_put() pair.
> If my understand was correct, __of_clk_get() uses __clk_create_clk(),
> so, its pair should be __clk_free_clk() ?
> I wonder I can find of_clk_get(), but couldn't find of_clk_put().
> Is using of_clk_get() <-> clk_put() OK ?
> 
> Best regards
> ---
> Kuninori Morimoto
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@...a-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

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