linux-kernel - Re: [PATCH 2/3] drivers: regulator: qcom: add PMS405 SPMI regulator

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]

Date:   Fri, 3 May 2019 10:29:42 +0200
From:   Jorge Ramirez <jorge.ramirez-ortiz@...aro.org>
To:     Mark Brown <broonie@...nel.org>
Cc:     lgirdwood@...il.com, robh+dt@...nel.org, mark.rutland@....com,
        linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        bjorn.andersson@...aro.org, vinod.koul@...aro.org,
        niklas.cassel@...aro.org, khasim.mohammed@...aro.org,
        linux-arm-kernel@...ts.infradead.org, linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH 2/3] drivers: regulator: qcom: add PMS405 SPMI regulator

On 5/3/19 08:26, Mark Brown wrote:
> On Thu, May 02, 2019 at 01:30:48PM +0200, Jorge Ramirez wrote:
>> On 5/2/19 04:33, Mark Brown wrote:
> 
>>> I'm not sure I follow here, sorry - I can see that the driver needs a
>>> custom get/set selector operation but shouldn't it be able to use the
>>> standard list and map operations for linear ranges?
> 
>> I agree it should, but unfortunately that is not the case; when I first
>> posted the patch I was concerned that for a regulator to be supported by
>> this driver it should obey to the driver's internals (ie: comply with
>> all of the spmi_common_regulator_registers definitions).
> 
> That's not a requirement that I'd particularly expect - it's not unusual
> for devices to have multiple different styles of regulators in a single
> chip (eg, DCDCs often have quite different register maps to LDOs).
> 
>> However, since there was just a single range to support, the
>> modifications I had to do to support this SPMI regulator were minimal -
>> hence why I opted for the changes under discussion instead of writing a
>> new driver (which IMO it is an overkill).
> 
>> what do you think?
> 
> It seems a bit of a jump to add a new driver - it's just another
> descriptor and ops structure isn't it?  Though as ever with the Qualcomm
> stuff this driver is pretty baroque which doesn't entirely help though I
> think it's just another regulator type which there's already some
> handling for.
> 

So how do we move this forward?

To sum up his regulator needs to be able to bypass accesses to
SPMI_COMMON_REG_VOLTAGE_RANGE and provide the range in some other way
hence the change below

I can't find a simpler solution than this since the function does now
what is supposed to do for all the regulator types supported in the driver

 @@ -653,6 +708,10 @@ spmi_regulator_find_range(struct spmi_regulator *vreg)
  	range = vreg->set_points->range;
  	end = range + vreg->set_points->count;

 +	/* we know we only have one range for this type */
 +	if (vreg->logical_type == SPMI_REGULATOR_LOGICAL_TYPE_HFS430)
 +		return range;
 +
  	spmi_vreg_read(vreg, SPMI_COMMON_REG_VOLTAGE_RANGE, &range_sel, 1);

  	for (; range < end; range++)