linux-kernel - Re: [PATCH] clk: rockchip: disable unused clocks

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 for Android: free password hash cracker in your pocket

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

Date:	Wed, 29 Oct 2014 20:05:49 +0100
From:	Heiko Stübner <heiko@...ech.de>
To:	Kever Yang <kever.yang@...k-chips.com>
Cc:	dianders@...omium.org, sonnyrao@...omium.org,
	addy.ke@...k-chips.com, cf@...k-chips.com, xjq@...k-chips.com,
	hj@...k-chips.com, dkl@...k-chips.com, huangtao@...k-chips.com,
	linux-rockchip@...ts.infradead.org,
	Mike Turquette <mturquette@...aro.org>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] clk: rockchip: disable unused clocks

Hi Kever,

Am Mittwoch, 29. Oktober 2014, 18:06:07 schrieb Kever Yang:
> The rockchip clock driver use CLK_IGNORE_UNUSED flag to make sure
> all the clocks are available like default power on state.
> We have implement the clock manage in most of rockchip drivers,
> it is time to remove it for power save.
> Instead we add CLK_IGNORE_UNUSED for some clock nodes which should
> be on during boot or no module driver in kernel will initialize it.
> 
> Signed-off-by: Kever Yang <kever.yang@...k-chips.com>

I did test it on all three socs I have and two issues popped up.

the easy issue:
- aclk_strc_sys in common_clk_branches
- aclk_core in rk3188_clk_branches
need to also stay on in clk-rk3188.c


the hard issue:
With this patch applied and the above fixed, both the Radxa Rock (rk3188) and 
Hayou Marsboard (rk3066a) produce imprecise external aborts when entering 
userspace.

I guess there is still a clock in there that should stay on. I've attached 
logs that highlight the clocks that get disabled in the late_initcall. [error 
happens with and without these extra messages].


I'll investigate further, but if you have an idea which of the clocks might be 
responsible I would be very glad :-)


thanks
Heiko

View attachment "screenlog.0.fail-rk3066a" of type "text/plain" (6721 bytes)

View attachment "screenlog.0.fail-rk3188" of type "text/plain" (9274 bytes)