linux-kernel - Re: [PATCH] ARM: OMAP2+: hwmod: Fix _wait_target

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

Date:	Tue, 13 Jan 2015 19:56:42 -0600
From:	Suman Anna <s-anna@...com>
To:	Paul Walmsley <paul@...an.com>
CC:	Lokesh Vutla <lokeshvutla@...com>, Roger Quadros <rogerq@...com>,
	<tony@...mide.com>, <santosh.shilimkar@...cle.com>,
	<t-kristo@...com>, <nm@...com>, <nsekhar@...com>,
	<bcousson@...libre.com>, <linux-omap@...r.kernel.org>,
	<linux-arm-kernel@...ts.infradead.org>,
	<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ARM: OMAP2+: hwmod: Fix _wait_target_ready() for hwmods
 without sysc

Hi Paul,

On 01/13/2015 05:29 PM, Paul Walmsley wrote:
> Hi Suman,
> 
> thanks for pitching in on this!
> 
> On Tue, 6 Jan 2015, Suman Anna wrote:
> 
>> You have removed the return from the above block on failure. If any DT
>> entry doesn't have the reg property, this will hang the kernel boot.
>> Just remove the "reg" entry from any of the existing DT, and you will
>> run into the issue, this is what 6423d6df1440 ("ARM: OMAP2+: hwmod:
>> check for module address space during init") fixed. 
> 
> Seems like that's the problem that we need to track down, then.  If a 
> hwmod has no MPU-accessible registers, it should still be possible to 
> init the clocks for the device, etc. ...

Yes true, and I should have rephrased above statement a little better -
its for modules with sysc but with no reg property to supply the base
for the module's SYSCONFIG or SYSSTATUS registers. The commit
6423d6df1440 has the explanation for the hang.

> 
>> Also, are you sure you want to turn the WARN into a pr_debug, it won't 
>> even show during the kernel boot log if the reg base is missing.
> 
> No, I'm not sure :-)  I guess it depends how many hwmods we'll have with 
> no MPU-accessible registers.  We don't seem to have address ranges for the 
> interconnects defined; we could fix that fairly easily.

The WARN_ON previously was to throw a eye-catchy print for the case
where hwmods have sysc but no address space defined (is an error
usually, but this is what we run into during the DT conversion of a
device as the hwmod and DTS changes come in through separate topic
branches).  I still think that the sysc check should be before the check
for _HWMOD_NO_MPU_PORT, a module with sysc mandates it has an MPU port.

regards
Suman
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/