linux-kernel - Re: [PATCH 08/13] coresight: Add support for CLAIM tag protocol

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] [day] [month] [year] [list]

Message-ID: <4aff79b4-593e-a19b-5e3f-de2a6d5bacb7@arm.com>
Date:   Thu, 16 Aug 2018 16:47:30 +0100
From:   Suzuki K Poulose <Suzuki.Poulose@....com>
To:     Mathieu Poirier <mathieu.poirier@...aro.org>
Cc:     linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        robert.walker@....com, mike.leach@....com
Subject: Re: [PATCH 08/13] coresight: Add support for CLAIM tag protocol

On 15/08/18 00:20, Mathieu Poirier wrote:
> On Mon, Aug 06, 2018 at 02:41:50PM +0100, Suzuki K Poulose wrote:
>> Add support for the CLAIM tag protocol for negotiating the
>> device ownership with other agents trying to use the coresight
>> component (internal vs. external). The Coresight architecture
>> specifies CLAIM tags (managed via CLAIMSET CLAIMCLR registers)
>> to negotiate the ownership of the device. PSCI recommends the
>> reservation of the bits in CLAIM tags for self-hosted and external
>> debug use. This patch implements the protocol for claiming
>> the devices before they are actually used.
> 
> I think the first paragraph of the cover letter (minus the reference to the
> documentation since you've included it below) would be perfect instead of the
> above.
> 
>>
>> Cc: Mathieu Poirier <mathieu.poirier@...aro.org>
>> Signed-off-by: Suzuki K Poulose <suzuki.poulose@....com>
>> ---
>>   drivers/hwtracing/coresight/coresight-priv.h |  7 +++
>>   drivers/hwtracing/coresight/coresight.c      | 85 ++++++++++++++++++++++++++++
>>   include/linux/coresight.h                    | 20 +++++++
>>   3 files changed, 112 insertions(+)

>> +void coresight_disclaim_device_unlocked(void __iomem *base)
>> +{
>> +
>> +	if (coresight_is_claimed_self_hosted(base))
>> +		coresight_clear_claim_tags(base);
>> +	else
>> +		/*
>> +		 * Either we or the external agent doesn't follow
>> +		 * the protocol.
>> +		 */
>> +		WARN_ON_ONCE(1);
> 
> When writing "... or the external agent doesn't follow the protocol", I deduce
> that an external agent would have trampled our claim tag.  I think this needs
> to be said explicitly in the comment.
> 

>> +static inline int coresight_claim_device_unlocked(void __iomem *base)
>> +{
>> +	return 0;
>> +}
>> +
>> +static inline int coresight_claim_device(void __iomem *base)
>> +{
>> +	return 0;
>> +}
> 
> Returning 0 would give a caller the impression the operation has succeeded when
> in fact it didn't.  I think we should return an error code here.


Agreed on all points, will respin.

Suzuki