linux-kernel - Re: [PATCH] musb: omap2430: do not assume balanced enable()/disable()

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]

Message-ID: <20160802170101.12bdad52@aktux>
Date:	Tue, 2 Aug 2016 17:01:01 +0200
From:	Andreas Kemnade <andreas@...nade.info>
To:	Tony Lindgren <tony@...mide.com>
Cc:	Bin Liu <b-liu@...com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	linux-usb@...r.kernel.org, linux-omap@...r.kernel.org,
	linux-kernel@...r.kernel.org, letux-kernel@...nphoenux.org
Subject: Re: [PATCH] musb: omap2430: do not assume balanced
 enable()/disable()

On Tue, 2 Aug 2016 03:33:34 -0700
Tony Lindgren <tony@...mide.com> wrote:

> * Andreas Kemnade <andreas@...nade.info> [160729 11:14]:
> > The code assumes that omap2430_musb_enable() and
> > omap2430_musb_disable() is called in a balanced way. The
> > That fact is broken by the fact that musb_init_controller() calls
> > musb_platform_disable() to switch from unknown state to off state.
> 
> OK, some spelling issues with the above paragraph though :)
> 
> > That means that phy_power_off() is called first so that
> > phy->power_count gets -1 and the phy is not enabled on
> > phy_power_on(). In the probably common case of using the
> > phy_twl4030, that prevents also charging the battery and so makes
> > further kernel debugging hard.
> 
> Is this with v4.7 kernel? Also, care to describe how you hit this
> and on which hardware? Just wondering..

I got this error on the Openphoenux GTA04 phone. It has a DM3730
SoC and a TPS65950 companion. Severe charging problems were already
observed with the 4.4rc1. I do not know if that already was exactly
*this* problem. I have debugged and patched the v4.7 kernel.
How I hit the problem: Just boot an that device and try to charge
via usb.
Should I resubmit the patch with an extended commit message?

Regards,
Andreas Kemnade

Content of type "application/pgp-signature" skipped