linux-kernel - Re: [PATCH] clk: qcom: smd-rpm: conditionally enable scaling before doing handover

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: <2a04cde4-ae8e-82b0-d467-0fc32d0c67b7@linaro.org>
Date:   Wed, 17 May 2023 21:11:41 +0200
From:   Konrad Dybcio <konrad.dybcio@...aro.org>
To:     Luca Weiss <luca@...tu.xyz>, ~postmarketos/upstreaming@...ts.sr.ht,
        phone-devel@...r.kernel.org,
        Bjorn Andersson <andersson@...nel.org>,
        Andy Gross <agross@...nel.org>,
        Michael Turquette <mturquette@...libre.com>,
        Stephen Boyd <sboyd@...nel.org>
Cc:     linux-arm-msm@...r.kernel.org, linux-clk@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] clk: qcom: smd-rpm: conditionally enable scaling before
 doing handover



On 17.05.2023 16:59, Luca Weiss wrote:
> On Dienstag, 16. Mai 2023 03:27:46 CEST Konrad Dybcio wrote:
>> On 6.05.2023 22:10, Luca Weiss wrote:
>>> On older platforms like msm8226, msm8974 and msm8916 the driver in the
>>> downstream kernel enables scaling first before doing the handover of the
>>> clocks.
>>>
>>> While this normally doesn't seem to cause noticeable problems, on
>>> apq8026-asus-sparrow this causes the device to immediately reboot,
>>> perhaps due to older rpm firmware that becomes unhappy.
>>>
>>> On newer platforms the order has swapped and enabling scaling is done
>>> after the handover, so let's introduce this behavior only conditionally
>>> for msm8226 and msm8974 for now.
>>>
>>> Signed-off-by: Luca Weiss <luca@...tu.xyz>
>>> ---
>>
>> Did you give this a spin on some 8974? I think hammerhead had
>> issues around rpmcc in the past..
> 
> Yes, appears to be fine on msm8974 also.
> 
> I tried to reproduce the hammerhead ocmem hang we had in the past but even 
> with v6.3 rpmcc it seems to be fine. But iirc it was happening more or less at 
> random in the past so wouldn't be surprised if my tests just didn't show 
> anything on accident.
> 
> Regards
> Luca
OK sg

Reviewed-by: Konrad Dybcio <konrad.dybcio@...aro.org>

Konrad
> 
>>
>> Konrad
>>
>>>  drivers/clk/qcom/clk-smd-rpm.c | 16 +++++++++++++---
>>>  1 file changed, 13 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/clk/qcom/clk-smd-rpm.c
>>> b/drivers/clk/qcom/clk-smd-rpm.c index 887b945a6fb7..6d5476afc4d1 100644
>>> --- a/drivers/clk/qcom/clk-smd-rpm.c
>>> +++ b/drivers/clk/qcom/clk-smd-rpm.c
>>> @@ -178,6 +178,7 @@ struct clk_smd_rpm_req {
>>>
>>>  struct rpm_smd_clk_desc {
>>>  
>>>  	struct clk_smd_rpm **clks;
>>>  	size_t num_clks;
>>>
>>> +	bool scaling_before_handover;
>>>
>>>  };
>>>  
>>>  static DEFINE_MUTEX(rpm_smd_clk_lock);
>>>
>>> @@ -693,6 +694,7 @@ static struct clk_smd_rpm *msm8974_clks[] = {
>>>
>>>  static const struct rpm_smd_clk_desc rpm_clk_msm8974 = {
>>>  
>>>  	.clks = msm8974_clks,
>>>  	.num_clks = ARRAY_SIZE(msm8974_clks),
>>>
>>> +	.scaling_before_handover = true,
>>>
>>>  };
>>>  
>>>  static struct clk_smd_rpm *msm8976_clks[] = {
>>>
>>> @@ -1318,6 +1320,12 @@ static int rpm_smd_clk_probe(struct platform_device
>>> *pdev)> 
>>>  	rpm_smd_clks = desc->clks;
>>>  	num_clks = desc->num_clks;
>>>
>>> +	if (desc->scaling_before_handover) {
>>> +		ret = clk_smd_rpm_enable_scaling(rpm);
>>> +		if (ret)
>>> +			goto err;
>>> +	}
>>> +
>>>
>>>  	for (i = 0; i < num_clks; i++) {
>>>  	
>>>  		if (!rpm_smd_clks[i])
>>>  		
>>>  			continue;
>>>
>>> @@ -1329,9 +1337,11 @@ static int rpm_smd_clk_probe(struct platform_device
>>> *pdev)> 
>>>  			goto err;
>>>  	
>>>  	}
>>>
>>> -	ret = clk_smd_rpm_enable_scaling(rpm);
>>> -	if (ret)
>>> -		goto err;
>>> +	if (!desc->scaling_before_handover) {
>>> +		ret = clk_smd_rpm_enable_scaling(rpm);
>>> +		if (ret)
>>> +			goto err;
>>> +	}
>>>
>>>  	for (i = 0; i < num_clks; i++) {
>>>  	
>>>  		if (!rpm_smd_clks[i])
>>>
>>> ---
>>> base-commit: dd9e11d6477a52ede9ebe575c83285e79e823889
>>> change-id: 20230506-rpmcc-scaling-handover-a63029ed9d13
>>>
>>> Best regards,
> 
> 
> 
>