Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <0c95c9a5-cf66-dcec-bfde-0ca201206c8b@linaro.org>
Date:   Tue, 15 Feb 2022 14:01:54 -0600
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.5.0
Subject: Re: [PATCH v3 01/25] bus: mhi: Fix pm_state conversion to string
Content-Language: en-US
To:     Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>,
        mhi@lists.linux.dev
Cc:     quic_hemantk@quicinc.com, quic_bbhatt@quicinc.com,
        quic_jhugo@quicinc.com, vinod.koul@linaro.org,
        bjorn.andersson@linaro.org, dmitry.baryshkov@linaro.org,
        quic_vbadigan@quicinc.com, quic_cang@quicinc.com,
        quic_skananth@quicinc.com, linux-arm-msm@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Paul Davey <paul.davey@alliedtelesis.co.nz>,
        Manivannan Sadhasivam <mani@kernel.org>,
        Hemant Kumar <hemantk@codeaurora.org>, stable@vger.kernel.org
References: <20220212182117.49438-1-manivannan.sadhasivam@linaro.org>
 <20220212182117.49438-2-manivannan.sadhasivam@linaro.org>
From:   Alex Elder <elder@linaro.org>
In-Reply-To: <20220212182117.49438-2-manivannan.sadhasivam@linaro.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2/12/22 12:20 PM, Manivannan Sadhasivam wrote:
> From: Paul Davey <paul.davey@alliedtelesis.co.nz>
> 
> On big endian architectures the mhi debugfs files which report pm state
> give "Invalid State" for all states.  This is caused by using
> find_last_bit which takes an unsigned long* while the state is passed in
> as an enum mhi_pm_state which will be of int size.

I think this would have fixed it too, but your fix is better.

	int index = find_last_bit(&(unsigned long)state, 32);

> Fix by using __fls to pass the value of state instead of find_last_bit.
> 
> Fixes: a6e2e3522f29 ("bus: mhi: core: Add support for PM state transitions")
> Signed-off-by: Paul Davey <paul.davey@alliedtelesis.co.nz>
> Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
> Reviewed-by: Hemant Kumar <hemantk@codeaurora.org>
> Cc: stable@vger.kernel.org
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> ---
>   drivers/bus/mhi/core/init.c | 8 +++++---
>   1 file changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/bus/mhi/core/init.c b/drivers/bus/mhi/core/init.c
> index 046f407dc5d6..af484b03558a 100644
> --- a/drivers/bus/mhi/core/init.c
> +++ b/drivers/bus/mhi/core/init.c
> @@ -79,10 +79,12 @@ static const char * const mhi_pm_state_str[] = {
>   
>   const char *to_mhi_pm_state_str(enum mhi_pm_state state)

The mhi_pm_state enumerated type is an enumerated sequence, not
a bit mask.  So knowing what the last (most significant) set bit
is not meaningful.  Or normally it shouldn't be.

If mhi_pm_state really were a bit mask, then its values should
be defined that way, i.e.,

	MHI_PM_STATE_DISABLE	= 1 << 0,
	MHI_PM_STATE_DISABLE	= 1 << 1,
	. . .

What's really going on is that the state value passed here
*is* a bitmask, whose bit positions are those mhi_pm_state
values.  So the state argument should have type u32.

This is a *separate* bug/issue.  It could be fixed separately
(before this patch), but I'd be OK with just explaining why
this change would occur as part of this modified patch.

>   {
> -	unsigned long pm_state = state;
> -	int index = find_last_bit(&pm_state, 32);
> +	int index;
>   
> -	if (index >= ARRAY_SIZE(mhi_pm_state_str))
> +	if (state)
> +		index = __fls(state);
> +
> +	if (!state || index >= ARRAY_SIZE(mhi_pm_state_str))
>   		return "Invalid State";

Do this test and return first, and skip the additional
check for "if (state)".

					-Alex

>   	return mhi_pm_state_str[index];