From mboxrd@z Thu Jan  1 00:00:00 1970
From: Stefan Wahren <stefan.wahren@i2se.com>
Date: Wed, 13 May 2015 10:14:19 +0200
Subject: [U-Boot] mxs: Possible in bug in mxs_enable_4p2_dcdc_input
In-Reply-To: <201505130913.25250.marex@denx.de>
References: <978811302.20941.1431496941951.JavaMail.open-xchange@oxbsltgw06.schlund.de>
 <201505130913.25250.marex@denx.de>
Message-ID: <555307DB.3090704@i2se.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Hi Marek,

Am 13.05.2015 um 09:13 schrieb Marek Vasut:
> On Wednesday, May 13, 2015 at 08:02:21 AM, Stefan Wahren wrote:
>> Hi,
> Hi,
>
>> during the study of DC-DC handling in u-boot i noticed a possible bug.
>>
>> In case the DC-DC is already enabled the function returns without
>> reenabling brown out detection. I don't think that's okay.
>> A possible fix would be to do the check if DC-DC is already enabled before
>> disabling brown out detection. This idea is
>> attached as a draft. If it's acceptable i will send a proper patch.
> What exactly would happen if you enable DCDC converter while BO detection
> is already enabled ? It is possible, that enabling the DCDC would cause a
> dip on the power rail and cause in turn a BO, no ? IIRC, that is the reason
> why BO detection is disabled when enabling DCDC.

i think we're talking about different things. The patch won't touch this
necessary behaviour.

No, i think it's faulty to leave the function with BO detection disabled
in the case the DC-DC is already enabled:

if (xfer && (readl(&power_regs->hw_power_5vctrl) &
                POWER_5VCTRL_ENABLE_DCDC)) {
         return;
}

>
> Best regards,
> Marek Vasut

Stefan