linux-kernel - Re: [PATCH 3/3] coresight: etm3x: Release CLAIM tag when operated from perf

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]

Message-ID: <054f82c1-451f-3c82-d5ab-e156c50e0c10@arm.com>
Date:   Thu, 8 Nov 2018 09:49:28 +0000
From:   Suzuki K Poulose <suzuki.poulose@....com>
To:     leo.yan@...aro.org, Mathieu Poirier <mathieu.poirier@...aro.org>
Cc:     linux-arm-kernel@...ts.infradead.org,
        alexander.shishkin@...ux.intel.com, coresight@...ts.linaro.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/3] coresight: etm3x: Release CLAIM tag when operated
 from perf

Leo,

On 07/11/2018 03:23, leo.yan@...aro.org wrote:
> Hi Mathieu,
> 
> On Mon, Nov 05, 2018 at 03:26:30PM -0700, Mathieu Poirier wrote:
>> This patch deals with the release of the CLAIM tag when the ETM is
>> operated from perf.  Otherwise the tag is left asserted and subsequent
>> requests to use the device fail.
>>
>> Signed-off-by: Mathieu Poirier <mathieu.poirier@...aro.org>
>> ---
>>   drivers/hwtracing/coresight/coresight-etm3x.c | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c
>> index fd5c4cca7db5..000796394662 100644
>> --- a/drivers/hwtracing/coresight/coresight-etm3x.c
>> +++ b/drivers/hwtracing/coresight/coresight-etm3x.c
>> @@ -603,6 +603,8 @@ static void etm_disable_perf(struct coresight_device *csdev)
>>   	 */
>>   	etm_set_pwrdwn(drvdata);
>>   
>> +	coresight_disclaim_device_unlocked(drvdata->base);
>> +


> 
> Just remind, this isn't consistent with the sequency in function
> etm_disable_hw(), which has the reversed sequence between
> etm_set_pwrdwn() and coresight_disclaim_device_unlocked().
> 
> Not sure which one sequence is more suitable, at the first glance,
> accessing register after pwrdwn related operation might have risk for
> deadlock?

Good point.

I assume that the CLAIMSET/CLR registers are in the same power domain as
the LAR (Software Lock Access register) accessed below. But I will
confirm this with the architect. Based on the response, we could
streamline both the sequences.

Suzuki

> 
> Thanks,
> Leo Yan
> 
>>   	CS_LOCK(drvdata->base);
>>   }
>>   
>> -- 
>> 2.7.4
>>