linux-kernel - Re: [PATCH V3 4/4] coresight: trbe: Enable ACPI based TRBE devices

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: <c9ba3dae-19ec-7ff7-6d07-e7b95f471954@arm.com>
Date:   Mon, 7 Aug 2023 17:28:00 +0530
From:   Anshuman Khandual <anshuman.khandual@....com>
To:     Suzuki K Poulose <suzuki.poulose@....com>,
        linux-arm-kernel@...ts.infradead.org
Cc:     Sami Mujawar <sami.mujawar@....com>,
        Catalin Marinas <catalin.marinas@....com>,
        Will Deacon <will@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Mike Leach <mike.leach@...aro.org>,
        Leo Yan <leo.yan@...aro.org>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        James Clark <james.clark@....com>, coresight@...ts.linaro.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH V3 4/4] coresight: trbe: Enable ACPI based TRBE devices



On 8/7/23 17:07, Suzuki K Poulose wrote:
> On 07/08/2023 05:43, Anshuman Khandual wrote:
>>
>>
>> On 8/3/23 11:26, Anshuman Khandual wrote:
>>> This detects and enables ACPI based TRBE devices via the dummy platform
>>> device created earlier for this purpose.
>>>
>>> Cc: Suzuki K Poulose <suzuki.poulose@....com>
>>> Cc: Mike Leach <mike.leach@...aro.org>
>>> Cc: Leo Yan <leo.yan@...aro.org>
>>> Cc: Alexander Shishkin <alexander.shishkin@...ux.intel.com>
>>> Cc: coresight@...ts.linaro.org
>>> Cc: linux-arm-kernel@...ts.infradead.org
>>> Cc: linux-kernel@...r.kernel.org
>>> Signed-off-by: Anshuman Khandual <anshuman.khandual@....com>
>>> ---
>>>   drivers/hwtracing/coresight/coresight-trbe.c | 9 +++++++++
>>>   drivers/hwtracing/coresight/coresight-trbe.h | 1 +
>>>   2 files changed, 10 insertions(+)
>>>
>>> diff --git a/drivers/hwtracing/coresight/coresight-trbe.c b/drivers/hwtracing/coresight/coresight-trbe.c
>>> index e1d9d06e7725..f884883e9018 100644
>>> --- a/drivers/hwtracing/coresight/coresight-trbe.c
>>> +++ b/drivers/hwtracing/coresight/coresight-trbe.c
>>> @@ -1537,7 +1537,16 @@ static const struct of_device_id arm_trbe_of_match[] = {
>>>   };
>>>   MODULE_DEVICE_TABLE(of, arm_trbe_of_match);
>>>   +#ifdef CONFIG_ACPI
>>> +static const struct platform_device_id arm_trbe_acpi_match[] = {
>>> +    { ARMV8_TRBE_PDEV_NAME, 0 },
>>> +    { }
>>> +};
>>> +MODULE_DEVICE_TABLE(platform, arm_trbe_acpi_match);
>>> +#endif
>>> +
>>>   static struct platform_driver arm_trbe_driver = {
>>> +    .id_table = arm_trbe_acpi_match,
>>
>> The build problem [1] reported on the first version of the series still exists
>> here i.e arm_trbe_acpi_match is hidden without CONFIG_ACPI. I had assumed that
>> CONFIG_CORESIGHT always enables CONFIG_ACPI, which is not the case. Following
>> random config (with CONFIG_ACPI=n and CONFIG_CORESIGHT_TRBE=y) easily triggers
>> the build problem.
>>
>> https://download.01.org/0day-ci/archive/20230805/202308052123.uqR35d19-lkp@intel.com/config
>>
>>   make CROSS_COMPILE=aarch64-linux-gnu- ARCH=arm64 -s -j 128
>> drivers/hwtracing/coresight/coresight-trbe.c:1563:23: error: implicit declaration of function ‘ACPI_PTR’ [-Werror=implicit-function-declaration]
>>   1563 |   .acpi_match_table = ACPI_PTR(arm_trbe_acpi_match),
>>        |                       ^~~~~~~~
>> drivers/hwtracing/coresight/coresight-trbe.c:1563:32: error: ‘arm_trbe_acpi_match’ undeclared here (not in a function); did you mean ‘arm_trbe_of_match’?
>>   1563 |   .acpi_match_table = ACPI_PTR(arm_trbe_acpi_match),
>>        |                                ^~~~~~~~~~~~~~~~~~~
>>        |                                arm_trbe_of_match
>>
>> Following config wrap around fixes the problem.
>>
>> --- a/drivers/hwtracing/coresight/coresight-trbe.c
>> +++ b/drivers/hwtracing/coresight/coresight-trbe.c
>> @@ -1557,7 +1557,9 @@ MODULE_DEVICE_TABLE(platform, arm_trbe_acpi_match);
>>   #endif
>>     static struct platform_driver arm_trbe_driver = {
>> +#ifdef CONFIG_ACPI
>>          .id_table = arm_trbe_acpi_match,
>> +#endif
>>          .driver = {
>>                  .name = DRVNAME,
>>                  .of_match_table = of_match_ptr(arm_trbe_of_match),
>>
>> Please not that unlike other coresight drivers, TRBE is not using 'acpi_device_id'
>> based "acpi_match_table = ACPI_PTR" construct. But regardless, ACPI_PTR() seems to
>> be an alternate (probably better) solution as well.
>>
>> --- a/drivers/hwtracing/coresight/coresight-trbe.c
>> +++ b/drivers/hwtracing/coresight/coresight-trbe.c
>> @@ -1557,7 +1557,7 @@ MODULE_DEVICE_TABLE(platform, arm_trbe_acpi_match);
>>   #endif
>>     static struct platform_driver arm_trbe_driver = {
>> -       .id_table = arm_trbe_acpi_match,
>> +       .id_table = ACPI_PTR(arm_trbe_acpi_match),
> 
> This is preferred.
> 
>>          .driver = {
>>                  .name = DRVNAME,
>>                  .of_match_table = of_match_ptr(arm_trbe_of_match),
>> diff --git a/drivers/hwtracing/coresight/coresight-trbe.h b/drivers/hwtracing/coresight/coresight-trbe.h
>> index 94e67009848a..fce1735d5c58 100644
>> --- a/drivers/hwtracing/coresight/coresight-trbe.h
>> +++ b/drivers/hwtracing/coresight/coresight-trbe.h
>> @@ -7,6 +7,7 @@
>>    *
>>    * Author: Anshuman Khandual <anshuman.khandual@....com>
>>    */
>> +#include <linux/acpi.h>
> 
> Shouldn't this be added in trbe.c ? Does trbe.h depend on any ACPI headers ?

The convention we have followed till now is all include/linux/ headers required
in TRBE driver is included via trbe.h not directly in trbe.c, just followed the
same principle this time around for acpi.h and perf/arm_pmu.h as well.